# US 2025/0164950 A1 · Building system with digital twin based agent processing

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0164950 A1
- Document ID: US20250164950A1
- Status: Published application
- Kind: A1
- Application number: 19/027,996
- Filed: 2025-01-17
- Published: 2025-05-22
- Earliest claimed: 2017-02-10
- Assignees: Johnson Controls Technology Company
- Applicants: Johnson Controls Technology Company
- Inventors: Youngchoon Park; Sudhi R. Sinha; Vaidhyanathan Venkiteswaran; Erik S. Paulson; Vijaya S. Chennupati; Kelsey C. Schuster
- Web page: https://patentgrep.com/patent/US20250164950A1

## CPC

- G05B 15/02: electric
- G06F 16/288: {Entity relationship models}
- G06N 3/006: based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
- G06N 5/043: Distributed expert systems; Blackboards
- G06N 20/00: Machine learning
- H04L 12/2809: {indicating that an appliance service is present in a home automation network (monitoring functionality H04L43/0817; discovery or management thereof, e.g. service location protocol [SLP] or web services, H04L67/51)}
- H04L 12/2827: {Reporting to a device within the home network; wherein the reception of the information reported automatically triggers the execution of a home appliance functionality}
- G05B 2219/2642: Domotique, domestic, home control, automation, smart house
- H04L 2012/285: {Generic home appliances, e.g. refrigerators}
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A system can include one or more memory devices that can store instructions thereon. The instructions can, when executed by one or more processors, cause the one or more processors to receive timeseries data associated with a building, detect that a new building device has been added to the building, determine that a representation of the new building device is absent from a digital twin of the building, and execute a machine learning model to add the representation of the new building device to the digital twin of the building.

## Claims

### Claim 1

A system comprising one or more memory devices storing instructions thereon that, when execute by one or more processors, cause the one or more processors to:
receive timeseries data associated with a building;
detect, responsive to receipt of the timeseries data, that a new building device has been added to the building;
determine that a representation of the new building device is absent from a digital twin of the building; and
execute a machine learning model to add the representation of the new building device to the digital twin of the building.

### Claim 2 (depends on 1)

The system of claim 1, wherein the instructions further case the one or more processors to:
execute the machine learning model to perform a device commissioning routine for the new building device, wherein performance of the device commissioning routine causes the one or more processors to:
generate, for the new building device, a device agent to provide the representation of the new building device within the digital twin;
map the new building device to a space of the building; and
register the device agent to a communication channel associated with the space of the building.

### Claim 3 (depends on 1)

The system of claim 1, wherein the new building device includes a device agent to provide the representation of the new building device within the digital twin, and wherein the instructions further cause the one or more processors to:
receive, from one or more agents registered to a communication channel, second timeseries data representative of one or more operations of the new building device; and
execute, responsive to receipt of the second timeseries data, the machine learning model to:
perform an operation on the device agent to simulate the one or more operations of the new building device;
determine, using (i) at least a portion of the second timeseries data and (ii) a result of the operation, a performance of the new building device; and
provide one or more or control signals to the new building device to adjust the performance of the new building device.

### Claim 4 (depends on 1)

The system of claim 1, wherein the instructions further cause the one or more processors to:
execute the machine learning model to perform a device monitoring routine, wherein performance of the device monitoring routine causes the one or more processors to:
subscribe to a communication channel;
receive, via the communication channel, second timeseries data associated with one or more operations of the new building device;
simulate, using one or more operational settings included in the second timeseries data, an operation of the new building device; and
determine a performance of the new building device based on (i) an outcome of simulating the operation and (ii) at least a portion of the second timeseries data.

### Claim 5 (depends on 4)

The system of claim 4, wherein performance of the device monitoring routine further causes the one or more processors to:
detect, responsive to determination of the performance of the new building device, that one or more aspects of the performance of the new building device violates a threshold;
generate one or more second operational settings for the new building device; and
provide, via the communication channel, the one or more second operational settings to the new building device to adjust the performance of the new building device.

### Claim 6 (depends on 4)

The system of claim 4, wherein the second timeseries data is produced as a result of a change to a status of a space of the building, wherein the one or more operational settings are configured to cause the new building device to address the change to the status of the space, and wherein performance of the device monitoring routine further causes the one or more processors to:
determine that the performance of the new building device conforms to a threshold based on the one or more operational settings having caused the new building device to address the change to the status of the space; and
maintain, in a database, a record which indicates that the change to the status of the space was addressed by the new building device operating in accordance with the one or more operational settings.

### Claim 7 (depends on 1)

The system of claim 1, wherein the new building device includes a device type, and wherein the instructions further cause the one or more processors to:
execute the machine learning model to perform a device commissioning routine for the new building device, wherein performance of the device commissioning routine causes the one or more processors to:
access a plurality of templates from a database, wherein each template of the plurality of templates represents a respective agent of a plurality of agents;
select, based on the device type, a first template of the plurality of templates; and
generate, based on an agent of the plurality of agents that is represented by the first template, a device agent for the new building device, the device agent to provide the representation of the new building device within the digital twin.

### Claim 8 (depends on 1)

The system of claim 1, wherein the instructions further cause the one or more processors to:
execute the machine learning model to perform a device commissioning routine for the new building device, wherein performance of the device commissioning routine causes the one or more processors to:
retrieve, from a database, one or more operational settings associated with a device type of the new building device;
provide, via a communication channel, the one or more operational settings to a device agent for the new building device; and
cause the new building device to operate in accordance with the one or more operational settings.

### Claim 9 (depends on 1)

The system of claim 1, wherein the instructions further cause the one or more processors to:
execute the machine learning model to:
receive, via a communication channel, second timeseries data representative of one or more operations of the new building device;
determine, responsive to receipt of the second timeseries data, that the new building device has experienced a fault; and
perform one or more operations on a device agent to address the fault experienced by the new building device.

### Claim 10

A method, comprising:
receiving, by one or more processing circuits, timeseries data associated with a building;
detecting, by the one or more processing circuits, responsive to receipt of the timeseries data, that a new building device has been added to the building;
determining, by the one or more processing circuits, that a representation of the new building device is absent from a digital twin of the building; and
executing, by the one or more processing circuits, a machine learning model to add the representation of the new building device to the digital twin of the building.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
executing, by the one or more processing circuits, the machine learning model to perform a device commissioning routine for the new building device, wherein performing the device commissioning routine includes:
generating, by the one or more processing circuits, for the new building device, a device agent to provide the representation of the new building device within the digital twin;
mapping, by the one or more processing circuits, the new building device to a space of the building; and
registering, by the one or more processing circuits, the device agent to a communication channel associated with the space of the building.

### Claim 12 (depends on 10)

The method of claim 10, wherein the new building device includes a device agent to provide the representation of the new building device within the digital twin, and further comprising:
receiving, by the one or more processing circuits, from one or more agents registered to a communication channel, second timeseries data representative of one or more operations of the new building device; and
executing, by the one or more processing circuits, responsive to receipt of the second timeseries data, the machine learning model to monitor a performance of the new building device, wherein monitoring the performance of the new building device includes:
performing, by the one or more processing circuits, an operation on the device agent to simulate the one or more operations of the new building device;
determining, by the one or more processing circuits, using (i) at least a portion of the second timeseries data and (ii) a result of the operation, the performance of the new building device; and
providing, by the one or more processing circuits, one or more or control signals to the new building device to adjust the performance of the new building device.

### Claim 13 (depends on 10)

The method of claim 10, further comprising:
executing, by the one or more processing circuits, the machine learning model to perform a device monitoring routine, wherein performing the device monitoring routine includes:
subscribing, by the one or more processing circuits, to a communication channel;
receiving, by the one or more processing circuits, via the communication channel, second timeseries data associated with one or more operations of the new building device;
simulating, by the one or more processing circuits, using one or more operational settings included in the second timeseries data, an operation of the new building device; and
determining, by the one or more processing circuits, a performance of the new building device based on (i) an outcome of simulating the operation and (ii) at least a portion of the second timeseries data.

### Claim 14 (depends on 13)

The method of claim 13, wherein performing the device monitoring routine further includes:
detecting, by the one or more processing circuits, responsive to determination of the performance of the new building device, that one or more aspects of the performance of the new building device violates a threshold;
generating, by the one or more processing circuits, one or more second operational settings for the new building device; and
providing, by the one or more processing circuits, via the communication channel, the one or more second operational settings to the new building device to adjust the performance of the new building device.

### Claim 15 (depends on 13)

The method of claim 13, wherein the second timeseries data is produced as a result of a change to a status of a space of the building, wherein the one or more operational settings are configured to cause the new building device to address the change to the status of the space, and wherein performing the device monitoring routine further includes:
determining, by the one or more processing circuits, that the performance of the new building device conforms to a threshold based on the one or more operational settings having caused the new building device to address the change to the status of the space; and
maintaining, by the one or more processing circuits, in a database, a record which indicates that the change to the status of the space was addressed by the new building device operating in accordance with the one or more operational settings.

### Claim 16 (depends on 10)

The method of claim 10, wherein the new building device includes a device type, and further comprising:
executing, by the one or more processing circuits, the machine learning model to perform a device commissioning routine for the new building device, wherein performing the device commissioning routine includes:
accessing, by the one or more processing circuits, a plurality of templates from a database, wherein each template of the plurality of templates represents a respective agent of a plurality of agents;
selecting, by the one or more processing circuits, based on the device type, a first template of the plurality of templates; and
generating, by the one or more processing circuits, based on an agent of the plurality of agents that is represented by the first template, a device agent for the new building device, the device agent to provide the representation of the new building device within the digital twin.

### Claim 17 (depends on 10)

The method of claim 10, further comprising:
executing, by the one or more processing circuits, the machine learning model to perform a device commissioning routine for the new building device, wherein performing the device commissioning routine includes:
retrieving, by the one or more processing circuits, from a database, one or more operational settings associated with a device type of the new building device;
providing, by the one or more processing circuits, via a communication channel, the one or more operational settings to a device agent for the new building device; and
causing, by the one or more processing circuits, the new building device to operate in accordance with the one or more operational settings.

### Claim 18

One or more non-transitory storage media storing instructions thereon that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving timeseries data associated with a building;
detecting, responsive to receipt of the timeseries data, that a new building device has been added to the building;
determining that a representation of the new building device is absent from a digital twin of the building; and
executing a machine learning model to add the representation of the new building device to the digital twin of the building.

### Claim 19 (depends on 18)

The one or more non-transitory storage media of claim 18, the operations further comprising:
executing the machine learning model to perform a device commissioning routine for the new building device, wherein performing the device commissioning routine includes:
generating, for the new building device, a device agent to provide the representation of the new building device within the digital twin;
mapping the new building device to a space of the building; and
registering the device agent to a communication channel associated with the space of the building.

### Claim 20 (depends on 18)

The one or more non-transitory storage media of claim 18, wherein the new building device includes a device agent to provide the representation of the new building device within the digital twin, and the operations further comprising:
receiving, from one or more agents registered to a communication channel, second timeseries data representative of one or more operations of the new building device; and
executing, responsive to receipt of the second timeseries data, the machine learning model to:
perform an operation on the device agent to simulate the one or more operations of the new building device;
determine, using (i) at least a portion of the second timeseries data and (ii) a result of the operation, a performance of the new building device; and
provide one or more or control signals to the new building device to adjust the performance of the new building device.
