IP Library Granted Patent US 10,481,874
Granted Patent B2
US 10,481,874 · App. 15/338,839 · Granted Nov 19, 2019

System architecture for secure and rapid development, deployment and management of analytics and software systems

Inventors: Arun Karthi Subramaniyan (Niskayuna, NY); John Lazos (Niskayuna, NY); Natarajan Chennimalai Kumar (Niskayuna, NY); Alexandre Iankoulski (Niskayuna, NY); Renato Giorgiani Do Nascimento (Rio de Janeiro, BR)
Assignee: GENERAL ELECTRIC COMPANY
G06F8/20G06F8/34G06F8/35G06F8/61G06F9/541
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,481,874
App. No.
15/338,839
Granted
Nov 19, 2019
Kind
B2
Abstract

According to some embodiments, system, apparatus and methods are provided comprising one or more component models of an analytic model for an installed product; an application programming interface (API) wrapper associated with each of the one or more component models, the API wrapper including information about one or more inputs to the component model; and wherein the component model and the API wrapper form a self-aware component. Numerous other aspects are provided.

Claims (56)

1. An apparatus for creating a prediction regarding operations of an installed product comprising:

one or more component models of an analytic model for the installed product, wherein the analytic model makes predictions of the installed product;

an application programming interface (API) wrapper associated with each of the one or more component models, the API wrapper including information about one or more inputs to the component model;

wherein each of the one or more component models and the associated API wrapper including a complete filesystem form a self-aware component; and

the analytical model, using one or more of the self-aware components, executes the same way regardless of an execution environment and creates the prediction regarding the installed product in the execution environment;

at least a communication channel to supply one or more outputs from the analytic model to the installed product to modify a condition of one or more physical elements of the installed product.

2. The apparatus of claim 1 , wherein information about the one or more inputs to the component model further comprises:

at least one source of the one or more input.

3. The apparatus of claim 2 , wherein the at least one input source is a file, a data stream, a data warehouse, a user, and a database.

4. The apparatus of claim 1 , wherein information about the one or more outputs from the component model further comprises:

at least one destination of the one or more outputs.

5. The apparatus of claim 4 , wherein at least one of the one or more outputs is stored in a JSON file.

6. The apparatus of claim 1 , wherein each component is operative to:

receive the one or more inputs from one or more sources;

execute the component model to generate one or more outputs; and

transmit the one or more outputs to at least one destination.

7. The apparatus of claim 1 , wherein a first component of the analytic model communicates with a second component of the analytic model via the API wrapper.

8. The apparatus of claim 1 , wherein information about the one or more inputs to the component model includes a definition of the input.

9. The apparatus of claim 1 , wherein each of the one or more components is self-registering.

10. The apparatus of claim 9 , wherein the self-registering component provides information about its contents, working rules and capabilities via emission of a unique signature to register itself in a system with valid interface and security settings.

11. The apparatus of claim 1 , wherein the API wrapper includes a modeling technique to create the analytic model.

12. The apparatus of claim 1 , wherein the component includes an integrated self-test for periodically checking the component model with new and existing datasets and trigger corrective actions to update the component model if the component model quality deviates beyond defined thresholds.

13. The apparatus of claim 1 , wherein metadata for the component is stored in the component when the component is running.

14. The apparatus of claim 1 , wherein if one component of the analytic model fails, the whole analytic model does not fail.

15. A method for creating a prediction regarding operations of an installed product comprising:

providing the installed product;

receiving one or more component models of an analytic model for the installed product, wherein the analytic model makes predictions of the installed product;

defining an AP wrapper associated with the component model, the definition including information about one or more inputs to the component and one or more outputs of the component model;

generating one or more self-aware components from the component model and the API wrapper including a complete filesystem;

using one or more of the self-aware components executing the analytical model the same way regardless of an execution environment and creates the prediction regarding the installed product in the execution environment; and

modifying a condition of one or more physical elements of the installed product based on an output signal from the analytic model.

16. The method of claim 15 , further comprising:

executing, via a computer processor, an analytic model for the installed product by executing each component.

17. The method of claim 16 , wherein executing each component further comprises:

executing a model task of each component model.

18. The method of claim 15 , wherein each component and the analytic model is self-registering.

19. The method of claim 15 , further comprising:

receiving information about the one or more inputs from at least one or more input sources.

20. The method of claim 15 , further comprising:

transmitting information about the one or more outputs from the component model to at least one destination.

21. The method of claim 15 , wherein a first component of the analytic model communicates with a second component of the analytic model via the API wrapper.

22. The method of claim 15 , wherein information about the one or more inputs to the component model includes a definition of the input.

23. A system for creating a prediction regarding operations of an installed product comprising:

one or more of the installed products;

a computer processor to:

receive a component model of an analytic model for the installed product, wherein the analytic model makes predictions of the installed product;

receive an application programming interface (API) wrapper associated with each of the one or more component models, the API wrapper including information about one or more inputs to the component model and one or more outputs of the component model, wherein the component model and the API wrapper including a complete filesystem form a self-aware component;

the analytical model, using one or more of the self-aware components, executes the same way regardless of an execution environment and creates the prediction regarding the installed product in the execution environment:

at least one communication channel for supplying one or more inputs to the component model;

a memory in communication with the computer processor, the memory storing the component and additional program instructions; and

wherein the communication channel supplies output from the analytic model to the installed product to modify a condition of one or more physical elements of the installed product.

24. The system of claim 18 , wherein the processor is operative with the additional program instructions to perform functions as follows:

receiving the one or more inputs from one or more input sources;

executing the component model to generate one or more outputs; and

transmitting the one or more outputs to at least one output destination.

25. The system of claim 18 , wherein each component and the analytic model is self-registering.

Assignments (4)
SECURITY INTEREST Recorded Mar 2, 2026
From: INNOVATEPRO MANAGEMENT USA LLC
To: ARES CAPITAL CORPORATION, AS COLLATERAL AGENT
Reel/Frame 073942/0369 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 27, 2026
From: GE VERNOVA ELECTRIFICATION SOFTWARE HOLDINGS LLC
To: INNOVATEPRO MANAGEMENT USA LLC
Reel/Frame 073924/0810 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 17, 2023
From: GENERAL ELECTRIC COMPANY
To: GE DIGITAL HOLDINGS LLC
Reel/Frame 065612/0085 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 31, 2016
From: SUBRAMANIYAN, ARUN KARTHI; LAZOS, JOHN; CHENNIMALAI KUMAR, NATARAJAN; IANKOULSKI, ALEXANDRE; DO NASCIMENTO, RENATO GIORGIANI
To: GENERAL ELECTRIC COMPANY
Reel/Frame 040175/0120 →
Continuity (1)
Related Publication 20180121170A1 · May 3, 2018