Fault diagnosis

• US 7,017,079 B2
• Filed: 08/27/2002
• Issued: 03/21/2006
• Est. Priority Date: 11/01/2001
• Status: Active Grant

• Current Assignee: Rolls-Royce PLC (Rolls-Royce Holdings Plc)
• Original Assignee: Rolls-Royce PLC (Rolls-Royce Holdings Plc)
• Inventors: Singh, Riti, Zedda, Marco, Gulati, Ankush
• Primary Examiner(s): Beausoliel, Robert W.
• Assistant Examiner(s): PUENTE, EMERSON C
• Application Number:
  US10/228,213
• Publication Number:
  US 20030088809A1
• Time in Patent Office: 1,302 Days
• Field of Search: 714/1, 714/25, 714/47, 702/185, 706 12- 14
• US Class Current: 714/25
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods

Abstract

The performance of each component in the system is defined by performance parameters x, which are related to measurement parameters z expressed as a function h( ) of the performance parameters x and operating parameters w. The method comprises: (a) setting an assumed maximum number of fault affected components and defining a series of fault classes corresponding to possible outcomes in terms of faulty components, (b) creating an initial population of strings for each fault class, each comprising a plurality of elements corresponding to the performance and operating parameters (c) optimising for each class an objective function J(x,w) which gives a measure of the consistency between measured values and calculated values of the measurement parameters calculated using said function of the performance parameters and the operating parameters h(x,w), and (d) selecting the class having the best value of the objective functions.

Claims

1. A computer implemented method for estimating the performance of components of a multi-component system,

the performance of each component in the system being defined by one or more performance parameters x, and

the performance parameters x being related to measurement parameters z which can be expressed as a function h( ) of the performance parameters x and operating parameters w defining an operating condition of the system;

the method comprising;

(a) setting an assumed maximum number of fault affected components and, based on this, defining a series of fault classes corresponding to possible outcomes in terms of faulty components,

(b) creating an initial population of strings for each fault class, each string comprising a plurality of elements corresponding to the performance parameters x and operating parameters w, values being assigned to the string elements which represent estimated values of said parameters and which are constrained only to indicate fault affected values for performance parameters defining the performance of the fault affected component or components of the respective class,

(c) optimising for each class an objective function J(x,w) which gives a measure of the consistency between measured values of the measurement parameters z collected from the system and calculated values of the measurement parameters calculated using said function of the performance parameters and the operating parameters h(x,w), and

(d) selecting the class or classes having the best value or values of the objective function;

said optimisation being performed for each class independently of the optimisations for the other classes and being an Evolutionary Program based method comprising the steps of;

(i) evaluating the objective function for each of the strings in the respective population and, on the basis of this evaluation, assigning a fitness value to each string, and

(ii) creating a next generation population of strings using a selection operator and at least one of a cross-over operator and a mutation operator, the selection operator employing the assigned fitness values to determine the strings that survive to the next generation, the cross-over operator, if used, being constrained to operate on pairs of strings from the same fault class, and the mutation operator, if used, being constrained so that it does not operate on string elements representing performance parameters other than those performance parameters associated with the fault affected components within the respective class.

2. A method according to claim 1, further comprising:

(e) employing a local search algorithm to further optimise the objective function of the or each class selected at step (d).

3. A method according to claim 1, further comprising employing at step (c) a local search algorithm to periodically refine the better strings in the population.

4. A method according to claim 1, further comprising:

selecting a maximum number M_bias of fault affected measurements, and

each time the objective function J(x,w) is evaluated during said optimisation, the method comprising calculating a plurality of possible values for the objective function, each possible value being calculated by determining the objective function with a different M_bias of the measurement parameters eliminated, the value of the minimum one of said plurality of calculated possible values being assigned as the value of the objective function.

5. A method according to claim 1, wherein:

said measurement parameters z are provided for two or more different operating points of said system, step (i) of said optimisation further comprising evaluating the objective function for each of said strings for each of said two or more operating points.

6. A method according to claim 1, wherein said function ho relating the measurement parameters to the performance and operating parameters is embodied in a performance simulation model designed to simulate the performance of the multi-component system to which the method is applied.

7. A computer implemented method for estimating the performance of components of a multi-component gas turbine engine,

the performance of each component in the gas turbine engine being defined by one or more performance parameters x, and

the performance parameters being related to the measurement parameters z which can be expressed as a function h( ) of the performance parameters x and operating parameters w defining an operating condition of the gas turbine engine;

the method comprising;

(a) setting an assumed maximum number of fault-affected components and, based on this, defining a series of fault classes corresponding to possible outcomes in terms of faulty components,

(b) optimising for each class independently of the optimisations for the other classes an objective function J(x,w) which gives a measure of the consistency between measured values of the measurement parameters z collected from the gas turbine engine and calculated values of the measurement parameters calculated using said function of the performance parameters and the operating parameters h(x,w), and

(c) selecting the class or classes having the best value or values of the objective function.

8. A method according to claim 7, wherein:

said measurement parameters z are provided for two or more different operating points of said gas turbine engine, said optimisation further comprising evaluating the objective function for each of said two or more operating points.

9. A method according to claim 7, wherein said function h( ) relating the measurement parameters to the performance and operating parameters is embodied in a performance simulation model designed to simulate the performance of the multi-component gas turbine engine to which the method is applied.

10. A computer system for estimating the performance of components of a multi-component system,

the performance of each component in the system being defined by one or more performance parameters x, and

the performance parameters x being related to measurement parameters z which can be expressed as a function h( ) of the performance parameters x and operating parameters w defining an operating condition of the multi-component system;

the computer system comprising;

input means for inputting values of the measurement parameters z collected from the multi-component system;

storage means for storing said collected values of the measurement parameters z and for storing an assumed maximum number of fault affected components;

processing means for (a), based on said assumed maximum number of fault affected components, defining a series of fault classes corresponding to possible outcomes in terms of faulty components, (b) creating an initial population of strings for each fault class, each string comprising a plurality of elements corresponding to the performance parameters x and operating parameters w, values being assigned to the string elements which represent estimated values of said parameters and which are constrained only to indicate fault affected values for performance parameters defining the performance of the fault affected component or components of the respective class, and (c) calculating values of the measurement parameters z using said function of the performance parameters and the operating parameters h(x,w); and

an optimiser which is adapted to (a) apply an Evolutionary Program based method to optimise for each class, independently of the optimisations for the other classes, an objective function J(x,w) which gives a measure of the consistency between said collected values of the measurement parameters z and said calculated values of the measurement parameters z, and (b) select the class or classes having the best value or values of the objective function,

said Evolutionary Program based method comprising the steps of;

(i) evaluating the objective function for each of the strings in the respective population and, on the basis of this evaluation, assigning a fitness value to each string, and

(ii) creating a next generation population of strings using a selection operator and at least one of a cross-over operator and a mutation operator, the selection operator employing the assigned fitness values to determine the strings that survive to the next generation, the cross-over operator, if used, being constrained to operate on pairs of strings from the same fault class, and the mutation operator, if used, being constrained so that it does not operate on string elements representing performance parameters other than those performance parameters associated with the fault affected components within the respective class.

11. A computer system according to claim 10, wherein the optimiser is further adapted to (c) employ a local search algorithm to further optimise the objective function of the or each selected class.

12. A computer system according to claim 10, wherein the optimiser is further adapted to employ a local search algorithm to periodically refine the better strings in the population.

13. A computer system according to claim 10, wherein:

the storage means further stores a maximum number M_bias of fault affected measurements, and

step (i) of the Evolutionary Program based method further comprises calculating a plurality of possible values for the objective function each time the objective function J(x,w) is evaluated, each possible value being calculated by determining the objective function with a different M_bias of the measurement parameters eliminated, the value of the minimum one of said plurality of calculated possible values being assigned as the value of the objective function.

14. A computer system according to claim 10, wherein said measurement parameters z are provided for two or more different operating points of said system, step (i) of the Evolutionary Program based method further comprising evaluating the objective function for each of said strings for each of said two or more operating points.

15. A computer system according to claim 10, wherein said to processing means comprises a performance simulation model for calculating values of the measurement parameters z, the performance simulation model being designed to simulate the performance of the multi-component system.

16. A computer system for estimating the performance of components of a multi-component gas turbine engine,

the performance of each component in the gas turbine engine being defined by one or more performance parameters x, and

the performance parameters x being related to measurement parameters z which can be expressed as a function h( ) of the performance parameters x and operating parameters w defining an operating condition of the gas turbine engine;

the computer system comprising;

input means for inputting values of the measurement parameters z collected from the gas turbine engine;

storage means for storing said collected values of the measurement parameters z and for storing an assumed maximum number of fault affected components;

processing means for (a), based on said assumed maximum number of fault affected components, defining a series of fault classes corresponding to possible outcomes in terms of faulty components, and (b) calculating values of the measurement parameters z using said function of the performance parameters and the operating parameters h(x,w); and

an optimiser which (a) optimises for each class, independently of the optimisations for the other classes, an objective function J(x,w) which gives a measure of the consistency between said collected values of the measurement parameters z and said calculated values of the measurement parameters z, and (b) selects the class or classes having the best value or values of the objective function.

17. A computer system according to claim 16, wherein said measurement parameters z are provided for two or more different operating points of said gas turbine engine, and said optimiser evaluates the objective function for each of said two or more operating points.

18. A computer system according to claim 16, wherein said processing means comprises a performance simulation model for calculating values of the measurement parameters z, the performance simulation model being designed to simulate the performance of the multi-component gas turbine engine.