BUILDING DATA PLATFORM WITH AI FEEDBACK ENRICHMENT

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

identify contextual data that describes an event in a virtual representation of a building;

enrich the event with the contextual data;

provide the event enriched with the contextual data to an artificial intelligence model, the artificial intelligence model trained to generate one or more predictions regarding at least one aspect of the building;

receive a first prediction generated by the artificial intelligence model using the event enriched with the contextual data; and

provide, to at least one of the artificial intelligence model or one or more second artificial intelligence models, the event enriched with the contextual data and the first prediction for subsequent generation of a second prediction.

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

provide the event enriched with the contextual data to a consuming system to cause the consuming system to generate a second event;

identify second contextual data that describes the second event in the virtual representation;

enrich the second event with the second contextual data; and

cause the artificial intelligence model to modify, using the second event enriched with the second contextual data, at least one aspect of the first prediction.

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

receive the second prediction generated by at least one of the artificial intelligence model or the one or more second artificial intelligence models;

identify one or more events in the virtual representation associated with the second prediction; and

enrich the one or more events with the second prediction.

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

receive one or more events from an event source;

enrich the one or more events with the first prediction;

identify, based on at least one subscription, a consuming system to execute one or more actions using the one or more events enriched with the first prediction; and

provide the one or more events enriched with the first prediction to the consuming system.

5 . The building system of claim 1 , wherein the virtual representation includes a digital twin, and wherein the instructions further cause the one or more processors to:

perform a first search of the digital twin to identify the contextual data; and

perform one or more second searches of the digital twin to identify second contextual data.

6 . The building system of claim 1 , wherein the contextual data includes at least one of:

a location within the building that an event source is located;

an indication of one or more pieces of building equipment of the event source; or

one or more capabilities associated with the one or more pieces of building equipment.

7 . The building system of claim 1 , wherein the event is received from an event source, and wherein the event source is at least one of:

an internal data source located within the building; or

an external data source located outside the building.

8 . The building system of claim 1 , wherein the virtual representation of the building includes a building graph, and wherein the instructions further cause the one or more processors to:

identify a first node of the building graph that represents an event source for the event; and

identify an edge of the building graph that connects the first node to a second node, and wherein the second node represents the contextual data.

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

execute a machine learning model using the first prediction to generate an inference of a characteristic of the building; and

identify, based on a subscription, a second machine learning model to execute on the inference of the characteristic of the building.

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

provide, responsive to receipt of the first prediction, the first prediction to the one or more second artificial intelligence models based at least on the one or more second artificial intelligence models having a subscription;

receive, from the one or more second artificial intelligence models, the second prediction; and

enrich one or more second events using the second prediction.

11 . A method, comprising:

identifying, by one or more processing circuits, contextual data that describes an event in a virtual representation of a building;

enriching, by the one or more processing circuits, the event with the contextual data;

providing, by the one or more processing circuits, the event enriched with the contextual data to an artificial intelligence model, the artificial intelligence model trained to generate one or more predictions regarding at least one aspect of the building;

receiving, by the one or more processing circuits, a first prediction generated by the artificial intelligence model using the event enriched with the contextual data; and

providing, by the one or more processing circuits, to at least one of the artificial intelligence model or one or more second artificial intelligence models, the event enriched with the contextual data and the first prediction for subsequent generation of a second prediction.

12 . The method of claim 11 , further comprising:

providing, by the one or more processing circuits, the event enriched with the contextual data to a consuming system to cause the consuming system to generate a second event;

identifying, by the one or more processing circuits, second contextual data that describes the second event in the virtual representation;

enriching, by the one or more processing circuits, the second event with the second contextual data; and

causing, by the one or more processing circuits, the artificial intelligence model to modify, using the second event enriched with the second contextual data, at least one aspect of the first prediction.

13 . The method of claim 11 , further comprising:

receiving, by the one or more processing circuits, the second prediction generated by at least one of the artificial intelligence model or the one or more second artificial intelligence models;

identifying, by the one or more processing circuits, one or more events in the virtual representation associated with the second prediction; and

enriching, by the one or more processing circuits, the one or more events with the second prediction.

14 . The method of claim 11 , further comprising:

receiving, by the one or more processing circuits, one or more events from an event source;

enriching, by the one or more processing circuits, the one or more events with the first prediction;

identifying, by the one or more processing circuits, based on at least one subscription, a consuming system to execute one or more actions using the one or more events enriched with the first prediction; and

providing, by the one or more processing circuits, the one or more events enriched with the first prediction to the consuming system.

15 . The method of claim 11 , wherein the virtual representation includes a digital twin, and further comprising:

performing, by the one or more processing circuits, a first search of the digital twin to identify the contextual data; and

performing, by the one or more processing circuits, one or more second searches of the digital twin to identify second contextual data.

16 . The method of claim 11 , wherein the contextual data includes at least one of:

a location within the building that an event source is located;

an indication of one or more pieces of building equipment of the event source; or

one or more capabilities associated with the one or more pieces of building equipment.

17 . The method of claim 11 , wherein the event is received from an event source, and wherein the event source is at least one of:

an internal data source located within the building; or

an external data source located outside the building.

18 . The method of claim 11 , wherein the virtual representation of the building includes a building graph, and further comprising:

identifying, by the one or more processing circuits, a first node of the building graph that represents an event source for the event; and

identifying, by the one or more processing circuits, an edge of the building graph that connects the first node to a second node, and wherein the second node represents the contextual data.

19 . The method of claim 11 , further comprising:

providing, by the one or more processing circuits, responsive to receiving the first prediction, the first prediction to the one or more second artificial intelligence models based at least on the one or more second artificial intelligence models having a subscription;

receiving, by the one or more processing circuits, from the one or more second artificial intelligence models, the second prediction; and

enriching, by the one or more processing circuits, one or more second events using the second prediction.

20 . A building system comprising:

one or more memory devices having instructions thereon; and

one or more processors configured to execute the instructions causing the one or more processors to:

receive, from an artificial intelligence model, a first prediction generated using one or more first events enriched with contextual data;

identify one or more second events or one or more second predictions associated with the first prediction;

enrich the one or more second events or the one or more second predictions with the first prediction; and

provide, based on a subscription, the one or more second events enriched with the first prediction or the one or more second predictions enriched with the first prediction.