System, device, and methods for updating system-monitoring models

• US 7,457,674 B2
• Filed: 08/24/2005
• Issued: 11/25/2008
• Est. Priority Date: 08/27/2004
• Status: Active Grant

• Current Assignee: Siemens Energy Incorporated (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG), Siemens Energy Incorporated (Siemens AG)
• Inventors: Neubauer, Claus, Yuan, Chao, Cataltepe, Zehra, Fang, Ming, Brummel, Hans-Gerd, McCorkle, Wesley
• Primary Examiner(s): DeCady; Albert
• Assistant Examiner(s): Garland; Steven R
• Application Number:
  US11/210,485
• Publication Number:
  US 20060047512A1
• Time in Patent Office: 1,189 Days
• Field of Search: 700/31, 700/29, 700/30, 700/47, 700/52
• US Class Current: 700/31
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 23/0243 model based detection method, e.g. first-principles knowledge model

Abstract

A system for updating a plurality of monitoring models is provided. The system includes a model association module that, for each of a plurality of monitored systems determines, an association between a particular monitored system and at least one of a plurality of estimation models. Each estimation model is based upon one of a plurality of distinct sets of estimation properties, and each set uniquely corresponds to a particular estimation model. The system also includes an updating module that updates at least one of the estimation properties and propagates the updated estimation properties to each estimation model that corresponds to a distinct set containing at least one estimation property that is updated. The system further includes a model modification module that modifies each estimation model that corresponds to a distinct set containing at least one estimation property that is updated.

Claims

1. An electronically-implemented model association method for updating estimation models used for system monitoring, the method comprising:

for each of a plurality of monitored systems, determining an association between a particular monitored system and at least one of a plurality of estimation models, wherein each estimation model is based upon one of a plurality of distinct sets of estimation properties, each set of estimation properties including at least one among a sensor list, sensor thresholds, training period, and estimation model algorithm, and wherein each set uniquely corresponds to a particular estimation model;

updating at least one of the estimation properties;

propagating the at least one updated estimation property to each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated;

modifying each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated;

updating an association between one or more estimation models and at least one of the plurality of monitored systems in response to a structure of at least one monitored system changing; and

retaining a system-specific version of each estimation model that is modified.

2. The method of claim 1, further comprising training a system-specific version of each estimation model for each monitored system, wherein the training is based upon sensor data generated by sensors communicatively linked to a particular system for which a particular system-specific version of an estimation model is trained.

3. The method of claim 2, further comprising retraining each system-specific version of each estimation model that is modified, wherein the retraining is based upon sensor data generated by sensors communicatively linked to a particular system for which a particular system-specific version of an estimation model is retrained.

4. The method of claim 3, wherein retraining is based upon a least-squares estimation.

5. The method of claim 1, further comprising generating a fleet table to represent the associations between the plurality of monitored systems and the plurality of estimation models.

6. The method of claim 5, wherein the fleet table comprises an M×

N matrix, M being an integer greater than one comprising a number of monitored systems, and N being an integer equal to a number of estimation models.

7. A system for updating a plurality of monitoring models, the system comprising:

a model association module that for each of a plurality of monitored systems determines an association between a particular monitored system and at least one of a plurality of estimation models, wherein each estimation model is based upon one of a plurality of distinct sets of estimation properties, each set of estimation properties including at least one among a sensor list, sensor thresholds, training period, and estimation model algorithm, and wherein each set uniquely corresponds to a particular estimation model;

an updating module that updates at least one of the estimation properties and propagates the at least one updated estimation property to each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated; and

a model modification module that modifies each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated;

the model modification module is configured to update an association between one or more estimation models and at least one of the plurality of monitored systems in response to a structure of at least one monitored system changing, and to cause a system-specific version of each estimation model that is modified to be electronically stored.

8. The system of claim 7, further comprising a training module configured to train a system-specific version of each estimation model for each monitored system based upon sensor data generated by sensors communicatively linked to a particular system for which a particular system-specific version of an estimation model is trained.

9. The system of claim 8, wherein the training module is further configured to retrain each system-specific version of each estimation model that is modified.

10. The system of claim 7, wherein the system is deployed at a location remote from the plurality of monitored systems.

11. A computer-readable storage medium comprising computer instructions for:

determining for each of a plurality of monitored systems an association between a particular monitored system and at least one of a plurality of estimation models, wherein each estimation model is based upon one of a plurality of distinct sets of estimation properties, each set of estimation properties including at least one among a sensor list, sensor thresholds, training period, and estimation model algorithm, and wherein each set uniquely corresponds to a particular estimation model;

updating at least one of the estimation properties;

propagating at least one updated estimation property to each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated;

modifying each estimation model that corresponds to a distinct set containing the at least one estimation property that is updated;

updating an association between one or more estimation models and at least one of the plurality of monitored systems in response to a structure of at least one monitored system changing; and

retaining a system-specific version of each estimation model that is modified.

12. The computer-readable storage medium of claim 11, further comprising computer instructions for training a system-specific version of each estimation model for each monitored system, wherein the training is based upon sensor data generated by sensors communicatively linked to a particular system for which a particular system-specific version of an estimation model is trained.

13. The computer-readable storage medium of claim 12, further comprising computer instructions for retraining each system-specific version of each estimation model that is modified, wherein the retraining is based upon sensor data generated by sensors communicatively linked to a particular system for which a particular system-specific version of an estimation model is retrained.

14. The computer-readable storage medium of claim 13, wherein the computer instructions for refraining comprise instructions for performing estimations based upon a least-squares method.