Monitoring and fault detection system and method using improved empirical model for range extrema

• US 7,373,283 B2
• Filed: 02/22/2001
• Issued: 05/13/2008
• Est. Priority Date: 02/22/2001
• Status: Active Grant

• Current Assignee: SmartSignal Corporation (Emerson Electric Company)
• Original Assignee: SmartSignal Corporation (Emerson Electric Company)
• Inventors: Wegerich, Stephan W., Herzog, James P.
• Primary Examiner(s): Rodriguez; Paul
• Assistant Examiner(s): Guill; Russ
• Application Number:
  US09/791,046
• Publication Number:
  US 20020152056A1
• Time in Patent Office: 2,637 Days
• Field of Search: 703/2, 706/26, 700/30, 700/31, 700/47
• US Class Current: 703/2
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04

Abstract

A method and apparatus for improved monitoring the operational state of instrumented systems is provided. An empirical model characterizes normal or desirable operation of the system, and real-time observations are provided to the model to generate estimates of expected sensor values. Comparison of the estimates with the real-time observations provides advanced warning of discrepancies in the operational state of the instrumented system. The invention provides for incipient failure detection, sensor failure detection and incipient process upset. An improved similarity operator provides for estimates that are not impaired by real-time observations at or beyond the limits of modeled data. The similarity operator comprises a difference function added to a constant, and the result is inverted.

Claims

1. A computable program product for computing similarity values for monitoring states of operation of a system, comprising a computer readable medium having instructions thereon which cause a processor to execute a process for:

selecting a next state vector from a reference library of state vectors comprising elemental parameter values from a memory;

retrieving a monitored state vector comprising elemental parameter values from an input source;

computing a difference function for pairs of corresponding elemental parameter values for a parameter of said system selected from the next state vector and the monitored state vector;

computing an elemental similarity s_c for pairs of corresponding elements, according to;

$s_c={\left(1+\left(\frac{{\mathrm{\theta <br><br>}}_c^{\mathrm{\lambda <br><br>}}}{\mathrm{\rho <br><br>}}\right)\right)}^{-1}$

for a corresponding pair of elements c where theta is the difference function, and lambda and rho are selectable constants;

combining elemental similarities for all pairs of corresponding elements to provide a vector similarity of the next state vector and the monitored state vector; and

storing in computer memory said vector similarity for use in providing information regarding states of operation of said system.

2. The computable program product of claim 1, wherein said instructions further cause a processor to execute a process for generating and storing in computer memory an estimated state vector from a composite of reference library state vectors using vector similarities of each state vector to the monitored state vector, said estimated state vector forming a basis for providing information regarding the states of operation of said system.

3. The computable program product of claim 1, wherein said instructions further cause a processor to execute a process for determining a highest vector similarity among vector similarities of each state vector in the reference library to the monitored state vector, and selecting and storing in computer memory a classification associated with the reference state vector having the highest vector similarity, said classification forming a basis for providing information regarding the states of operation of said system.

4. The computable program product of claim 1, wherein said instructions for causing a processor to execute a process for combining elemental similarities for all pairs of corresponding elements to provide a vector similarity of the next state vector and the monitored state vector, compute the vector similarity S_v as:

$S_V=\frac{1}{N}<br><br>\underset{c=1}{\overset{N}{\sum <br><br>}}<br><br>{\left(1+\left(\frac{<br><br>Y_c-R_c<br><br>}{{\mathrm{\rho <br><br>}}_c}\right)\right)}^{-1}$

for monitored state vector Y and next state vector R, each having at least N elements, where Y_c and R_c are the C^th elements of Y and R respectively, and rho-sub-c (ρ

_c) is an expected range for the C^th elements.

5. A method for empirically modeling multiple parameters of a system, wherein said system is selected from a process, a machine or a biological system, the empirical model used to analyze said system, said method comprising:

providing a reference library of snapshots of time-related values of said parameters representing operation of said system;

providing a monitored snapshot of values of at least some of said parameters;

computing for the at least some parameters values of a difference function of corresponding parameters of said monitored snapshot and said reference library snapshots;

computing for the at least some parameters elemental similarities of corresponding parameters of said monitored snapshot and a reference library snapshot as the inverse of;

a constant plus the value of the difference function,

computing a snapshot similarity from the elemental similarities of corresponding parameters of the monitored snapshot and a reference library snapshot, and

storing in a computer memory at least one said snapshot similarity for use in an analysis of said system.

6. The method of claim 5, further comprising the step of generating and storing in a computer memory an estimate snapshot of said parameters based on at least one snapshot similarity and using said estimate snapshot in said analysis.

7. The method of claim 6, further comprising the step of comparing the estimate snapshot to the monitored snapshot and generating and storing in a computer memory an alert on determining a difference.

8. The method of claim 5, further comprising:

providing a classification associated with at least some of the reference library snapshots; and

determining and storing in a computer memory a classification for said monitored snapshot based on at least one snapshot similarity and the reference library snapshot classifications.

9. The method of claim 8, wherein said determining step comprises determining the class of the reference library snapshot having the highest snapshot similarity with said monitored snapshot.

10. The method of claim 5, wherein said difference function is a function of the larger of two corresponding parameters minus the smaller of the two corresponding parameters, divided by an expected range for such parameters.

11. The method of claim 10, wherein said snapshot similarity (S_v) is computed according to:

$S_V=\frac{1}{N}<br><br>\underset{c=1}{\overset{N}{\sum <br><br>}}<br><br>{\left(1+\left(\frac{<br><br>Y_c-R_c<br><br>}{{\mathrm{\rho <br><br>}}_c}\right)\right)}^{-1}$

for monitored snapshot Y and reference library snapshot R, each having at least N corresponding parameters, where Y_c and R_c are the C^th parameters of Y and R respectively, and rho-sub-c (ρ

_c) is an expected range for the C^th parameters.

12. The method of claim 5, wherein said difference function is a function of the difference between a trigonometric function of the first of two corresponding parameters and a trigonometric function of the second of the two corresponding parameters.

13. An apparatus for monitoring the state of a system instrumented with sensors, wherein said system is selected from a process or a machine, said apparatus comprising:

a memory for storing a reference library of snapshots of time-related values of said sensors representing operation of said system;

means for acquiring a monitored snapshot of sensor values from the system;

a similarity module implemented in software and executed in a processor that receives the monitored snapshot from said means for acquiring and generates similarity values for comparisons of the monitored snapshot to reference library snapshots in said memory, for providing an indication of the state of the system based on the similarity values; and

output means that stores in a computer memory a system state based on the similarity values;

where each similarity value for a comparison of the monitored snapshot to a reference library snapshot is a statistical composite of similarity values for corresponding elements of the monitored snapshot and the reference library snapshot, wherein each similarity value for corresponding elements is derived from the inverse of a quantity comprising a constant plus a difference of the corresponding elements, said difference divided by an expected range for said elements.

14. The apparatus of claim 13, further comprising an estimated state generator implemented in software and executed in a processor, that generates a snapshot of estimated sensor values, based on the generated similarity values, said estimated sensor values forming a basis for indicating the state of said system.

15. The apparatus of claim 14, wherein the estimated state generator generates the estimated sensor values as a composite of the reference library snapshots, weighted by corresponding similarity values from said similarity module.

16. The apparatus of claim 14, further comprising a deviation detection module implemented in software and executed in a processor, that receives the estimated sensor values from said estimated state generator and the monitored snapshot, and generates an indication of a difference between them, as a basis for indicating the state of said system.

17. The apparatus of claim 16, wherein the deviation detection module differences the estimated sensor values with the monitored snapshot to produce a residual snapshot, and performs a sequential probability ratio test on at least one sensor across a sequence of such residual snapshots, as a basis for indicating the state of said system.

18. The apparatus of claim 13, wherein said output means stores in a computer memory a classification associated with at least one reference library snapshot, based on which reference library snapshot has the highest similarity value with the monitored snapshot.

19. The apparatus of claim 13, wherein the statistical composite of similarity values for corresponding elements of the monitored snapshot and the reference library snapshot is calculated according to:

$S=\frac{\underset{c=1}{\overset{N}{\sum <br><br>}}<br><br>S_c}{N}$

where s_c is the similarity value for corresponding elements c of snapshots comprising N elements.

20. The apparatus of claim 13, wherein said difference of the corresponding elements is equal to the absolute value of the difference of the corresponding elements.

21. The apparatus of claim 13, wherein said difference of the corresponding elements is equal to the difference of a trigonometric function of each of the corresponding elements individually.