Method and apparatus for improved fault detection in power generation equipment

• US 20060074595A1
• Filed: 08/12/2005
• Published: 04/06/2006
• Est. Priority Date: 08/25/2004
• Status: Active Grant

• Current Assignee: Siemens Corp. (Siemens AG)
• Original Assignee: Siemens Corp. (Siemens AG)
• Inventors: Neubauer, Claus, Yuan, Chao, Cataltepe, Zehra
• Granted Patent: US 7,953,577 B2
• Time in Patent Office: Days
• US Class Current: 702/182
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05B 23/0297 Reconfiguration of monitoring system, e.g. use of virtual sensors; change monitoring method as a response to monitoring results

Abstract

A method and apparatus for detecting faults in power plant equipment is discloses using sensor confidence and an improved method of identifying the normal operating range of the power generation equipment as measured by those sensors. A confidence is assigned to a sensor in proportion to the residue associated with that sensor. If the sensor has high residue, a small confidence is assigned to the sensor. If a sensor has a low residue, a high confidence is assigned to that sensor, and appropriate weighting of that sensor with other sensors is provided. A feature space trajectory (FST) method is used to model the normal operating range curve distribution of power generation equipment characteristics. Such an FST method is illustratively used in conjunction with a minimum spanning tree (MST) method to identify a plurality of nodes and to then connect those with line segments that approximate a curve.

Claims

11. An apparatus for improving operational data measurements in equipment having a monitoring system, said monitoring system comprising a plurality of sensors, said apparatus comprising:

means for receiving an element of operational data from a sensor in said plurality of sensors, said element of operational data representing a measurement of an observed condition of said apparatus;

means for determining a confidence level for said sensor using said element of operational data; and

means for determining an estimate of said measurement as a function of said confidence level.

12. The apparatus of claim 11 wherein said equipment comprises power generation equipment.

13. The apparatus of claim 11 wherein said means for determining a confidence level comprises means for calculating the expression:

14. The apparatus of claim 11 wherein said means for determining an improved estimate comprises means for calculating the expression:

{circumflex over (x)}_iw_i·

y_i+(1−

w_i)·

{tilde over (x)}_i

where w_i is a confidence of the i-th sensor;

{circumflex over (x)}_i is an improved estimate of said measurement;

{tilde over (x)}_i is an estimate of a sensor vector from the i-th sensor; and

y_i is the observed sensor vector at the i-th sensor.

15. The apparatus of claim 11 wherein said means for determining a confidence level comprises:

means for calculating a normalized absolute difference between an observed sensor value and an estimated sensor value {tilde over (x)}; and

means for mapping said normalized absolute difference to a confidence level.

16. The apparatus of claim 15 wherein said means for mapping comprises means for mapping said normalized absolute difference to a confidence level function, said function defined by the expression:

g(d)=exp(γ

d²)

where d is the normalized absolute difference; and

γ

is a selected convergence factor.

17. The apparatus of claim 16 wherein γ

is selected in a way such that g(1)=0.001.

18. The apparatus of claim 15 wherein said means for calculating a normalized absolute difference between an observed sensor value and an estimated sensor value {tilde over (x)} comprises:

means for determining a plurality of nodes in a set of training data, said nodes representing said normal operating range; and

means for estimating a curve representing a normal operating range of said apparatus by connecting each node in said plurality of nodes with at least one other node in a way such that the sum of the length of said connections is a minimum.

19. The apparatus of claim 18 wherein said estimated sensor value x is a vector located on said curve.

20. A method for improving operational data measurements in an apparatus having a monitoring system, said monitoring system adapted to compare at least a first observed value associated with one operating characteristic of said apparatus to a normal operating range of that characteristic, said method comprising:

determining a plurality of nodes from a set of training data, said nodes representing said normal operating range of said apparatus; and

connecting each node in said plurality of nodes with at least one other node in a way such that the sum of the length of said connections is a minimum.

21. The method of claim 20 wherein said step of determining comprises identifying a plurality of centroids in a set of training data using a k-mean clustering algorithm.

22. The method of claim 21 wherein said step of determining comprises determining the distance from each element of data in said training data to at least one centroid.

23. The method of claim 22 wherein said step of determining comprises forming a plurality of nodes by associating each element of data to that centroid to which said element is closest.

24-1. A monitoring system for detecting fault conditions in equipment comprising:

a first sensor in a plurality of sensors for collecting an element of operational data said element of operational data representing a measurement of an observed condition of said equipment;

means for determining a confidence level for said sensor using said element of operational data; and

means for determining an estimate of said measurement as a function of said confidence level 25. The monitoring system of claim 24 wherein said equipment comprises power generation equipment.