Granted Patent B2

US 12,406,193 · App. 18/372,469 · Granted Sep 2, 2025

Inventors: Rajiv Ramanasankaran (San Jose, CA); Young M. Lee (Old Westbury, NY)

Assignee: TYCO FIRE & SECURITY GMBH

G06N5/025G06F16/9024G06F16/9035G06F16/906G06N5/022

Abstract

Claims (88)

1. A building system comprising one or more storage devices storing instructions thereon that, when executed by one or more processors, cause the one or more processors to:

store a data structure of a digital twin of an entity of a building, the data structure storing information associated with at least one of the entity or a plurality of entities of the building;

execute a learning algorithm to construct a trigger rule and an action rule with the data structure of the digital twin as an input to the learning algorithm;

store the trigger rule and the action rule in the digital twin;

determine that the trigger rule of the digital twin is triggered by comparing at least some of the information of the data structure against one or more conditions of the trigger rule by querying the data structure for data values and comparing the data values against the one or more conditions of the trigger rule;

determine an action of the action rule of the digital twin responsive to determining that the trigger rule is triggered by executing the action rule; and

cause one or more devices to operate based on the action determined by the action rule.

2. The building system of claim 1 , wherein the instructions cause the one or more processors to:

provide the action to one or more other digital twins;

determine that the action triggers one or more trigger rules of the one or more other digital twins; and

determine one or more actions of one or more action rules of the one or more other digital twins by executing the one or more action rules responsive to the action triggering the one or more trigger rules.

3. The building system of claim 1 , wherein the data structure is a graph data structure comprising a plurality of nodes and a plurality of edges between the plurality of nodes, wherein the plurality of nodes represent the entity, the plurality of entities, and data points of the entity and the plurality of entities;

wherein the plurality of edges represent relationships between the entity, the plurality of entities, and the data points of the entity and the plurality of entities.

4. The building system of claim 1 , wherein the instructions cause the one or more processors to cause the one or more devices to operate based on the action by:

sending a command to the one or more devices to perform the action; or

store the action in the data structure and retrieve the action from the data structure and send the command based on the action retrieved from the data structure to the one or more devices.

5. The building system of claim 1 , wherein the instructions cause the one or more processors to learn one or more parameters of the trigger rule or the action rule by:

adjusting a value of the one or more parameters of the trigger rule or the action rule to a plurality of different values;

simulating one or more states resulting from the plurality of different values of the one or more parameters of the trigger rule or the action rule; and

determine one or more updated values for the one or more parameters of the trigger rule or the action rule based on the one or more states resulting from the plurality of different values of the one or more parameters of the trigger rule or the action rule.

6. The building system of claim 5 , wherein the instructions cause the one or more processors to determine the one or more updated values for the one or more parameters of the trigger rule or the action rule by:

determining a particular reward for the one or more states resulting from the plurality of different values of the one or more parameters of the trigger rule or the action rule;

training a model that predicts a reward based on the particular reward for the one or more states and the plurality of different values of the one or more parameters of the trigger rule or the action rule; and

performing an optimization that determines the one or more updated values for the one or more parameters of the trigger rule or the action rule based on the model and the reward.

7. The building system of claim 6 , wherein the instructions cause the one or more processors to generate construct the trigger rule and the action rule based on actions and the information associated with the entity and actions and the information associated with the plurality of entities by:

determining trigger rules for the information associated with the entity and the information associated with the plurality of entities;

determining action rules for the actions of the entity and the actions of the plurality of entities;

simulating combinations of the trigger rules and the action rules, and determining rewards for the combinations; and

selecting the trigger rule and the action rule from the combinations of the trigger rules and the action rules based on the rewards of the combinations.

8. The building system of claim 6 , wherein the instructions cause the one or more processors to construct the trigger rule and the action rule based on actions and the information associated with the entity and actions and the information associated with the plurality of entities by:

determining trigger rules for the information associated with the entity and the information associated with the plurality of entities, the trigger rules including the trigger rule;

determining action rules for the actions of the entity and the actions of the plurality of entities, the action rules including the action rule;

monitoring behavior of the building system of the building;

determining a relationship between the trigger rule of the trigger rules and the action rule of the action rules indicated by the behavior of the building system; and

selecting the trigger rule from the trigger rules and the action rule from the action rules based on the relationship.

9. A method comprising:

storing, by a processing circuit, a data structure of a digital twin of an entity, the data structure storing information associated with at least one of the entity or a plurality of entities;

executing, by the processing circuit, a learning algorithm to construct a trigger rule and an action rule with the data structure of the digital twin as an input to the learning algorithm;

storing, by the processing circuit, the trigger rule and the action rule in the digital twin;

determining, by the processing circuit, that the trigger rule of the digital twin is triggered by comparing at least some of the information of the data structure against one or more conditions of the trigger rule by querying the data structure for data values and comparing the data values against the one or more conditions of the trigger rule;

determining, by the processing circuit, an action of the action rule of the digital twin responsive to determining that the trigger rule is triggered by executing the action rule; and

causing, by the processing circuit, one or more devices to operate based on the action determined by the action rule.

10. The method of claim 9 , further comprising:

providing, by the processing circuit, the action to one or more other digital twins;

determining, by the processing circuit, that the action triggers one or more trigger rules of the one or more other digital twins; and

determining, by the processing circuit, one or more actions of one or more action rules of the one or more other digital twins by executing the one or more action rules responsive to the action triggering the one or more trigger rules.

11. The method of claim 9 , wherein the data structure is a graph data structure comprising a plurality of nodes and a plurality of edges between the plurality of nodes, wherein the plurality of nodes represent the entity, the plurality of entities, and data points of the entity and the plurality of entities;

wherein the plurality of edges represent relationships between the entity, the plurality of entities, and the data points of the entity and the plurality of entities.

12. The method of claim 9 , wherein causing, by the processing circuit, the one or more devices to operate based on the action comprises:

sending a command to the one or more devices to perform the action; or

storing the action in the data structure and retrieve the action from the data structure and send the command based on the action retrieved from the data structure to the one or more devices.

13. The method of claim 9 , further comprising learning, by the processing circuit, one or more parameters of the trigger rule or the action rule by:

adjusting a value of the one or more parameters of the trigger rule or the action rule to a plurality of different values;

simulating one or more states resulting from the plurality of different values of the one or more parameters of the trigger rule or the action rule; and

determining one or more updated values for the one or more parameters of the trigger rule or the action rule based on the one or more states resulting from the plurality of different values of the one or more parameters of the trigger rule or the action rule.

14. The method of claim 9 , further comprising generating, by the processing circuit, the digital twin by:

identifying actions and data associated with the entity based on the data structure;

identifying that the plurality of entities are related to the entity based on the data structure;

identifying actions and data associated with the plurality of entities based on the data structure; and

generating the trigger rule and the action rule based on the actions and the data associated with the entity and the actions and the data associated with the plurality of entities.

15. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

storing a data structure of a digital twin of an entity of a building, the data structure storing information associated with at least one of the entity or a plurality of entities of the building;

executing a learning algorithm to construct a trigger rule and an action rule with the data structure of the digital twin as an input to the learning algorithm;

storing the trigger rule and the action rule in the digital twin;

determining that the trigger rule of the digital twin is triggered by comparing at least some of the information of the data structure against one or more conditions of the trigger rule by querying the data structure for data values and comparing the data values against the one or more conditions of the trigger rule;

determining an action of the action rule of the digital twin responsive to determining that the trigger rule is triggered by executing the action rule; and

causing one or more devices to operate based on the action determined by the action rule.

16. The one or more non-transitory computer-readable media of claim 15 , the operations further comprising:

identifying that a relationship exists between the trigger rule and the action rule; and

determining the action of the action rule responsive to determining that the trigger rule is triggered and that the relationship exists between the trigger rule and the action rule.

17. The one or more non-transitory computer-readable media of claim 15 , the operations further comprising:

storing a second digital twin associated with a second entity of the building, the second digital twin comprising a second action rule;

identifying a relationship between the trigger rule of the digital twin and the action rule of the second digital twin; and

determining a second action based on the second action rule of the second digital twin responsive to determining that the trigger rule of the digital twin is triggered.

18. The one or more non-transitory computer-readable media of claim 15 , the operations further comprising:

receiving first timeseries data from the entity;

receiving second timeseries data from a second entity of the building separate from the entity; and

determining that the trigger rule of the digital twin is triggered by comparing the first timeseries data and the second timeseries data against the one or more conditions of the trigger rule.

19. The one or more non-transitory computer-readable media of claim 15 , the operations to generate the digital twin further comprising:

identifying actions and data associated with the entity based on the data structure;

identifying that the plurality of entities are related to the entity based on the data structure;

identifying actions and data associated with the plurality of entities based on the data structure; and

generating the trigger rule and the action rule based on the actions and the data associated with the entity and the actions and the data associated with the plurality of entities.

20. The one or more non-transitory computer-readable media of claim 15 , the operations further cause the one or more processors to construct the trigger rule and the action rule based on actions and the information associated with the entity and actions and the information associated with the plurality of entities by:

determining trigger rules for the information associated with the entity and the information associated with the plurality of entities;

determining action rules for the actions of the entity and the actions of the plurality of entities;

simulating combinations of the trigger rules and the action rules, and determining rewards for the combinations; and

selecting the trigger rule and the action rule from the combinations of the trigger rules and the action rules based on the rewards of the combinations.

Assignments (2)

Continuity (2)

References Cited (400)