Method and system to model industrial assets using heterogenous data sources

Patent metadata

Field	Value
Publication number	US-10928811-B2
Application number	US-201715793590-A
Country	US
Kind code	B2
Filing date	Oct 25, 2017
Priority date	Oct 25, 2017
Publication date	Feb 23, 2021
Grant date	Feb 23, 2021

Abstract

According to some embodiments, a system and method are provided to model a sparse data asset. The system comprises a processor and a non-transitory computer-readable medium comprising instructions that when executed by the processor perform a method to model a sparse data asset. Relevant data and operational data associated with the newly operational are received. A transfer model based on the relevant data and the received operational data. An input into the transfer model is received and a predication based on data associated with the received operational data and the relevant data is output.

First claim

What is claimed:

1. A system to model a sparse data asset, the system comprising:
a processor; and
a non-transitory computer-readable medium comprising instructions that when executed by the processor perform a method to model a sparse data asset, the instructions to:
receive relevant data associated with the sparse data asset, wherein the relevant data is a physics model associated with a peer asset currently in operation;
receive operational data associated with the sparse data asset;
create, via the processor, a transfer model based on the relevant data and the received operational data, wherein the creating of the transfer model comprises:
creating an initial model based on the received relevant data;
applying the initial model to received operational data;
determining an error rate associated with target data;
creating a second model based on the error rate associated with the target data;
determining output data of the second model as second source data;
determining a second error rate for the second source data as corrected source data; and
creating the transfer model based on the corrected source data;

receive an input into the transfer model; and
output a predication based on the input, data associated with the received operational data, and the relevant data.

2. The system of claim 1 , wherein the relevant data comprises non-asset specific simulation data or operation data from peer assets.

3. The system of claim 1 , wherein the relevant data is determined based on a specific analytics task of interest.

4. The system of claim 1 , wherein the peer asset is connected to the sparse data asset via a network for transmitting the relevant data.

5. A non-transitory computer-readable medium comprising instructions that when executed by the processor perform a method to model a sparse data asset, the method comprising:
receiving relevant data associated with the sparse data asset, wherein the relevant data is a physics model associated with a peer asset currently in operation;
receiving operational data associated with the asset;
creating, via a processor, a transfer model based on the relevant data and the received operational data, wherein the creating of the transfer model comprises:
creating an initial model based on received source data from the relevant data;
applying the initial model to received operational data;
determining an error rate associated with target data;
creating a second model based on the error rate associated with the target data;
determining output data of the second model as second source data;
determining a second error rate for the second source data as corrected source data; and
creating the transfer model based on the corrected source data;

receiving an input into the transfer model; and
outputting a predication based on the input, data associated with the received operational data, and the relevant data.

6. The non-transitory computer-readable medium of claim 5 , wherein the relevant data comprises non-asset specific simulation or operation data.

7. The non-transitory computer-readable medium of claim 5 , wherein the relevant data is determined based on a specific analytics task of interest.

8. The non-transitory computer-readable medium of claim 5 , wherein the peer asset is connected to the sparse data asset via a network for transmitting the relevant data.

9. A method to model a sparse data asset, the method comprising:
receiving relevant data associated with the sparse data asset, wherein the relevant data is a physics model associated with a peer asset currently in operation;
receiving operational data associated with the asset;
creating, via a processor, a transfer model based on the relevant data and the received operational data, wherein the creating of the transfer model comprises:
creating an initial model based on received source data from the relevant data;
applying the initial model to received operational data;
determining an error rate associated with target data;
creating a second model based on the error rate associated with the target data;
determining output data of the second model as second source data;
determining a second error rate for the second source data as corrected source data; and
creating the transfer model based on the corrected source data;

receiving an input into the transfer model; and
outputting a predication based on the input, data associated with the received operational data, and the relevant data.

10. The method of claim 9 , wherein the relevant data comprises non-asset specific simulation or operation data.

11. The method of claim 9 , wherein the relevant data is determined based on a specific analytics task of interest.

12. The method of claim 9 , wherein the peer asset is connected to the sparse data asset via a network for transmitting the relevant data.

Assignees

• Gen Electric

Inventors

• Xin Fuxiao
• Swanson Larry
• Xu Rui
• Salter Morgan
• Pandey Achalesh
• Chandra Ramu
• Yan Weizhong

Classifications

• F03D80/50

  Maintenance or repair · CPC title

• G06Q50/06

  Energy or water supply · CPC title

• G05B23/0251 Primary

  Abstraction hierarchy, e.g. "complex systems", i.e. system is divided in subsystems, subsystems are monitored and results are combined to decide on status of whole system · CPC title

• Y02E10/72

  Wind turbines with rotation axis in wind direction · CPC title

• G06N3/02

  Neural networks · CPC title