Cloud-based fire protection system and method

Patent metadata

Field	Value
Publication number	US-11520946-B2
Application number	US-201916379374-A
Country	US
Kind code	B2
Filing date	Apr 9, 2019
Priority date	Apr 9, 2019
Publication date	Dec 6, 2022
Grant date	Dec 6, 2022

Abstract

A system performs cloud-based fire protection. The system receives, by a cloud platform, data from one or more initiating devices. The system stores the data in a persistent data storage of the cloud platform over a period of time. The system applies machine learning to the data to build or adjust a predictive detection model. The system processes, by computing resources of the cloud platform, the data using the predictive detection model to determine an existence of a safety event. The system then transmits, to at least one notification device, an event notification in response to the existence of the safety event.

First claim

What is claimed is:

1. A method of cloud-based fire protection, comprising:
receiving, by a cloud platform, data from one or more fire protection devices, wherein the one or more fire protection devices are configured to communicate with the cloud platform to collectively provide a fire alarm system associated with a geographical or logical system;
storing the data in a persistent data storage of the cloud platform over a period of time;
applying machine learning to the data received from the one or more fire protection devices and to other data received from other fire protection devices configured in a different geographical or logical system, to extract a pattern indicative of a type of fire being threatening or non-threatening;
adjusting a fire event detection threshold of a predictive detection model based on the pattern;
processing, by computing resources of the cloud platform, the data using the predictive detection model to determine an existence of a safety event; and
transmitting, to at least one notification device, an event notification in response to the existence of the safety event.

2. The method of claim 1 , wherein the processing of the data comprises:
comparing the data against an event threshold or identifying a signal pattern of the data; and
determining the existence of the safety event in response to the comparing or the identifying.

3. The method of claim 1 , wherein the one or more fire protection devices communicate with the cloud platform over the Internet.

4. The method of claim 1 , wherein the cloud platform communicates with the at least one notification device over the Internet.

5. The method of claim 1 , wherein the at least one notification device comprises a remote annunciator.

6. The method of claim 1 , wherein the at least one notification device comprises a personal computer or a personal mobile device.

7. The method of claim 1 , wherein the event notification comprises at least one of an email, a text message, or a mobile push notification.

8. The method of claim 1 , wherein the receiving of the data comprises receiving raw sensor data from the one or more fire protection devices.

9. The method of claim 8 , wherein the receiving of the data further comprises receiving processed data generated by processing the raw sensor data by the one or more fire protection devices or by one or more hubs, routers, gateways, or a combination thereof, that couple the one or more fire protection devices with the cloud platform.

10. The method of claim 1 , further comprising:
monitoring, by the cloud platform, a connectivity of the one or more fire protection devices and the at least one notification device with the cloud platform; and
generating, by the cloud platform, a lost connectivity indication when the connectivity is lost.

11. The method of claim 10 , wherein the lost connectivity indication comprises one or more of an audible alarm, a visual alarm, a text message, or an email.

12. The method of claim 10 , wherein the monitoring comprises:
determining, by the cloud platform, the connectivity based on a regularity of reception of the data by the cloud platform from the one or more fire protection devices.

13. The method of claim 10 , wherein the monitoring comprises:
determining, by the cloud platform, the connectivity based on a periodic status report received by the cloud platform from the one or more fire protection devices or from the at least one notification device.

14. The method of claim 1 , further comprising:
receiving, via an interface of the cloud platform, a user input indicating an invalidation of the safety event; and
adjusting a corresponding event detection threshold in the predictive detection model in response to the invalidation.

15. The method of claim 1 , wherein the one or more fire protection devices are logically grouped together in a zone, a building floor, a building section, a building, a building complex, or a campus.

16. The method of claim 1 , wherein the one or more fire protection devices are configured to communicate with the cloud platform to collectively provide the fire alarm system without a need for association with a local fire panel control unit.

17. The method of claim 1 , wherein the fire alarm system is a de-centralized cloud-based fire protection system that includes a local fire alarm control panel, wherein the local fire alarm control panel is configured as a gateway to the Internet to couple the one or more fire protection devices with the cloud platform such that the cloud platform processes business logic of the fire alarm system.

18. A system for cloud-based fire protection, comprising:
one or more processors providing computing resources for a cloud platform; and
a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to:
receive, by the cloud platform, data from one or more fire protection devices, wherein the one or more fire protection devices are configured to communicate with the cloud platform to collectively provide a fire alarm system associated with a geographical or logical system;
store the data in a persistent data storage of the cloud platform over a period of time;
apply machine learning to the data received from the one or more fire protection devices and to other data received from other fire protection devices configured in a different geographical or logical system, to extract a pattern indicative of a type of fire being threatening or non-threatening;
adjust a fire event detection threshold of a predictive detection model based on the pattern;
process the data using the predictive detection model to determine an existence of a safety event; and
transmit, to at least one notification device, an event notification in response to the existence of the safety event.

19. The system of claim 18 , wherein the one or more fire protection devices and the at least one notification device communicate with the cloud platform over the Internet.

20. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors providing computing resources for a cloud platform, cause the one or more processors to:
receive, by the cloud platform, data from one or more fire protection devices, wherein the one or more fire protection devices are configured to communicate with the cloud platform to collectively provide a fire alarm system associated with a geographical or logical system;
store the data in a persistent data storage of the cloud platform over a period of time;
apply machine learning to the data received from the one or more fire protection devices and to other data received from other fire protection devices configured in a different geographical or logical system, to extract a pattern indicative of a type of fire being threatening or non-threatening;
adjust a fire event detection threshold of a predictive detection model based on the pattern;
process the data using the predictive detection model to determine an existence of a safety event; and
transmit, to at least one notification device, an event notification in response to the existence of the safety event.

Assignees

• Johnson Controls Fire Prot Lp
• Johnson Controls Tyco IP Holdings LLP

Inventors

• Tran Liem-Binh
• Bourbonnais David-Alexandre

Classifications

• G06F30/00 Primary

  Computer-aided design [CAD] · CPC title

• G08B17/00

  Fire alarms; Alarms responsive to explosion · CPC title

• G08B31/00 Primary

  Predictive alarm systems characterised by extrapolation or other computation using updated historic data · CPC title

• G06F9/542

  Event management; Broadcasting; Multicasting; Notifications · CPC title

• G08B26/007

  Wireless interrogation · CPC title