Granted Patent B1

US 12,265,366 · App. 18/244,779 · Granted Apr 1, 2025

Inventors: Michael J. Risbeck (Madison, WI); Shawn D. Schubert (Oak Creek, WI); Jonathan D. Douglas (Mequon, WI); Charles Gans (Phoenixville, PA); Bernard P. Clement (Mequon, WI); Young M. Lee (Old Westbury, NY); Edward Gerard McNamara (Old Pallas, IE); Brennan H. Fentzlaff (Oconomowoc, WI)

Assignee: TYCO FIRE & SECURITY GMBH

G05B19/042G05B2219/25011

Abstract

Claims (50)

1. A method for controlling building equipment, the method comprising:

generating building data comprising synthetic measurements of one or more conditions of a building space using a generative artificial intelligence model;

determining whether the building data correspond to conditions in the building space that comply with one or more regulations or certification standards by comparing a value of an environmental condition or an operating parameter for the building equipment of the building data with a threshold for the environmental condition or a threshold for the operating parameter for the building equipment set by the one or more regulations or certifications standards; and

in response to determining that the building data correspond to conditions of the building space that comply with the one or more regulations or certification standards, using the building data to train another model used to manage or control the building equipment or the building space.

2. The method of claim 1 , wherein using the building data to operate the building equipment comprises:

using the building data to train a predictive model or reinforcement policy; and

using the predictive model to generate operating parameters for the building equipment.

3. The method of claim 1 , further comprising in response to determining that the building data correspond to conditions in the building space that do not comply with the one or more regulations or certification standards:

filtering the building data to remove one or more portions of the building data that correspond to conditions in the building space that do not comply with the one or more regulations or certification standards; and

initiating a fine-tuning process of the generative artificial intelligence model based on the filtered building data.

4. The method of claim 1 , further comprising:

generating a compliance report that indicates one or more of a plurality of regulations or certification standards with which the building data comply, the building data including design parameters for a new building; and

operating a display screen to present the compliance report to a user.

5. The method of claim 1 , wherein the one or more regulations or certifications standards comprise one or more comfort standards, one or more emissions standards, one or more American Society of Heating, Refrigerating, and Air-Conditioning Engineers (“ASHRAE”) standards, and an International WELL Building Institute standard.

6. The method of claim 1 , wherein the building data comprise time-series data of operating conditions of the building equipment over a time period.

7. The method of claim 1 , wherein the generative artificial intelligence model is trained based on a database of real-world building data of a plurality of different buildings.

8. A control system for building equipment, the control system comprising:

building equipment that serve a building space; and

processing circuitry configured to:

generate building data comprising synthetic measurements of one or more conditions of the building space using a generative artificial intelligence model;

determine whether the building data correspond to conditions in the building space that comply with one or more regulations or certification standards by comparing a value of an environmental condition or an operating parameter for the building equipment of the building data with a threshold for the environmental condition or a threshold for the operating parameter for the building equipment set by the one or more regulations or certifications standards; and

in response to determining that the building data correspond to conditions of the building space that comply with the one or more regulations or certification standards, using the building data to train another model to manage or control the building equipment or the building space.

9. The control system of claim 8 , wherein the processing circuitry is configured to use the building data to operate the building equipment by:

using the building data to train a predictive model or reinforcement policy; and

using the predictive model to generate operating parameters for the building equipment.

10. The control system of claim 8 , wherein the processing circuitry is further configured to, in response to determining that the building data correspond to conditions in the building space that do not comply with the one or more regulations or certification standards:

filter the building data to remove one or more portions of the building data that correspond to conditions in the building space that do not comply with the one or more regulations or certification standards; and

initiate a fine-tuning process of the generative artificial intelligence model based on the filtered building data.

11. The control system of claim 8 , wherein the processing circuitry is further configured to:

generate a compliance report that indicates one or more of a plurality of regulations or certification standards with which the building data comply, the building data including design parameters for a new building; and

operate a display screen to present the compliance report to a user.

12. The control system of claim 8 , wherein the one or more regulations or certifications standards comprise one or more comfort standards, one or more emissions standards, one or more American Society of Heating, Refrigerating, and Air-Conditioning Engineers (“ASHRAE”) standards, and an International WELL Building Institute standard.

13. The control system of claim 8 , wherein the building data comprise time-series data of operating conditions of the building equipment over a time period.

14. The control system of claim 8 , wherein the generative artificial intelligence model is trained based on a database of real-world building data of a plurality of different buildings.

15. A method for controlling building equipment, the method comprising:

generating building data comprising operating parameters for building equipment using a generative artificial intelligence model;

predicting a condition of a building space that will result from operating the building equipment according to the operating parameters using a predictive model;

determining whether the condition of the building space complies with one or more regulations or certification standards by comparing the condition of the building space with a threshold set by the one or more regulations or certification standards;

in response to determining that the condition complies with the one or more regulations or certification standards, using the building data to train another model to manage or control the building equipment or the building space.

16. The method of claim 15 , wherein using the building data further comprises:

using the building data to train a reinforcement policy; and

using the other model to generate operating parameters for the building equipment.

17. The method of claim 15 , further comprising in response to determining that the building data correspond to conditions in the building space that do not comply with the one or more regulations or certification standards:

filtering the building data to remove one or more portions of the building data that correspond to conditions in the building space that do not comply with the one or more regulations or certification standards; and

initiating a fine-tuning process of the generative artificial intelligence model based on the filtered building data.

18. The method of claim 15 , further comprising:

generating a compliance report that indicates one or more of a plurality of regulations or certification standards with which the building data comply, the building data including design parameters for a new building; and

operating a display screen to present the compliance report to a user.

19. The method of claim 15 , wherein the one or more regulations or certifications standards comprise one or more comfort standards, one or more emissions standards, one or more American Society of Heating, Refrigerating, and Air-Conditioning Engineers (“ASHRAE”) standards, and an International WELL Building Institute standard.

20. The method of claim 15 , wherein the generative artificial intelligence model is trained based on a database of real-world building data of a plurality of different buildings and wherein the generative artificial intelligence model is a large language model.

Assignments (2)

References Cited (32)