Diagnostic and prognostic method and system

• US 7,499,777 B2
• Filed: 04/08/2005
• Issued: 03/03/2009
• Est. Priority Date: 04/08/2005
• Status: Active Grant

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Willden, Wade, Grichnik, Anthony J., Seskin, Michael
• Primary Examiner(s): Beaulieu; Yonel
• Application Number:
  US11/101,544
• Publication Number:
  US 20060230313A1
• Time in Patent Office: 1,425 Days
• Field of Search: 701/29, 701 33- 34, 714 25- 26, 702182-185
• US Class Current: 701/29
• 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: G07C 5/0808 Diagnosing performance data testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005

Abstract

A method for diagnosing a system is provided. The method may include obtaining a plurality of input parameters from the system. The system may generate actual values of one or more output parameters based on the plurality of input parameters. The method may also include independently deriving values of the one or more output parameters, based on the plurality of input parameters, using a computational model, and detecting a divergence between the derived values and the actual values. Further, the method may include identifying respective desired statistic distributions of the plurality of input parameters, based on the actual values of the one or more output parameters, using the computational model.

Claims

1. A method for diagnosing a system, comprising:

obtaining a plurality of input parameters from the system, wherein the system generates actual values of one or more output parameters based on the plurality of input parameters;

independently deriving values of the one or more output parameters, based on the plurality of input parameters, using a computational model;

detecting a divergence between the derived values and the actual values;

identifying respective desired statistical distributions of the plurality of input parameters, based on the actual values of the one or more output parameters, using the computational model;

determining whether any of the plurality of input parameters is out of a predetermined input space prior to identifying the respective desired statistical distributions, wherein the input space is determined based on original desired statistical distributions of the plurality of input parameters of the computational model; and

determining the derived values of the one or more output parameters inconclusive for identifying potential failures if any of the plurality of input parameters is out of the input space.

2. The method according to claim 1, further including:

identifying a potential failure associated with one or more of the input parameters by comparing the identified desired statistical distributions with original desired statistical distributions of the plurality of input parameters of the computational model.

3. The method according to claim 2, further including:

performing a predetermined diagnostic action corresponding to the potential failure.

4. The method according to claim 1, wherein the identifying further includes:

determining a candidate set of input parameters with a maximum zeta statistic using a genetic algorithm; and

determining the desired statistical distributions of the plurality of input parameters based on the candidate set,

wherein the zeta statistic ζ

is represented by;

$\mathrm{\zeta <br><br>}=\stackrel{j}{\underset{1}{\sum <br><br>}}<br><br>\stackrel{i}{\underset{1}{\sum <br><br>}}<br><br><br><br>S_{\mathrm{ij}}<br><br><br><br>\left(\frac{{\mathrm{\sigma <br><br>}}_i}{{\stackrel{\_}{x}}_i}\right)<br><br>\left(\frac{{\stackrel{\_}{x}}_j}{{\mathrm{\sigma <br><br>}}_j}\right),$

provided that x_i represents a mean of an ith input parameter from the plurality of input parameters;

x_j represents a mean of a jth output parameter from the one or more output parameters;

σ

_i represents a standard deviation of the ith input parameter from the plurality of input parameters;

σ

_j represents a standard deviation of the jth output parameter from the one or more output parameters; and

|S_ij| represents sensitivity of the jth output parameter to the ith input parameter of the computational model.

5. The method according to claim 1, wherein the computational model is generated by:

obtaining data records associated with the plurality of input parameters and the one or more output parameters;

selecting a subset of input parameters from the plurality of input parameters;

generating the computational model indicative of interrelationships between the subset of the plurality of input parameters and the one or more output parameters based on the data records; and

determining desired respective statistical distributions of the subset of input parameters of the computational model based on a zeta statistic of the computational model.

6. The method according to claim 5, wherein selecting further includes:

pre-processing the data records; and

using a genetic algorithm to select the subset from the plurality of input parameters based on a mahalanobis distance between a normal data set and an abnormal data set of the data records.

7. The method according to claim 5, wherein generating further includes:

creating a neural network computational model;

training the neural network computational model using the data records; and

validating the neural network computation model using the data records.

8. The method according to claim 5, wherein determining further includes:

determining a candidate set of input parameters with a maximum zeta statistic using a genetic algorithm; and

determining the desired statistical distributions of the input parameters based on the candidate set,

wherein the zeta statistic ζ

is represented by;

$\mathrm{\zeta <br><br>}=\stackrel{j}{\underset{1}{\sum <br><br>}}<br><br>\stackrel{i}{\underset{1}{\sum <br><br>}}<br><br><br><br>S_{\mathrm{ij}}<br><br><br><br>\left(\frac{{\mathrm{\sigma <br><br>}}_i}{{\stackrel{\_}{x}}_i}\right)<br><br>\left(\frac{{\stackrel{\_}{x}}_j}{{\mathrm{\sigma <br><br>}}_j}\right),$

provided that x_i represents a mean of an ith input;

x_j represents a mean of a jth output;

σ

_i represents a standard deviation of the ith input;

σ

_j represents a standard deviation of the jth output; and

|S_ij| represents sensitivity of the jth output to the ith input of the computational model.

9. A method for providing a prognosis associated with a system, comprising:

obtaining a plurality of input parameters from the system, wherein the system generates actual values of one or more output parameters based on the plurality of input parameters;

independently deriving values of the one or more output parameters using a computational model based on the plurality of input parameters;

estimating future values of the one or more output parameters of the system by calculating a trend of differences between the actual values and the derived values;

detecting a potential divergence between the future values and the derived values; and

identifying respective desired statistic distributions of the plurality of input parameters using the computational model based on the future values of the one or more output parameters,

wherein the computational model is generated by;

obtaining data records associated with the plurality of input parameters and the one or more output parameters;

generating the computational model indicative of interrelationships between the plurality of input parameters and the one or more output parameters based on the data records; and

determining original desired respective statistical distributions of the subset of input parameters of the computational model based on a zeta statistic of the computational model.

10. The method according to claim 9, further including:

predicting a potential failure associated with one or more input parameters by comparing the identified desired statistical distributions with original desired statistical distributions of the plurality of input parameters of the computational model.

11. The method according to claim 10, further including:

notifying a user of the system of the potential failure.

12. The method according to claim 9, wherein the identifying further includes:

determining a candidate set of input parameters with a maximum zeta statistic using a genetic algorithm; and

determining the desired statistical distributions of the plurality input parameters based on the candidate set,

wherein the zeta statistic ζ

is represented by;

$\mathrm{\zeta <br><br>}=\stackrel{j}{\underset{1}{\sum <br><br>}}<br><br>\stackrel{i}{\underset{1}{\sum <br><br>}}<br><br><br><br>S_{\mathrm{ij}}<br><br><br><br>\left(\frac{{\mathrm{\sigma <br><br>}}_i}{{\stackrel{\_}{x}}_i}\right)<br><br>\left(\frac{{\stackrel{\_}{x}}_j}{{\mathrm{\sigma <br><br>}}_j}\right),$

provided that x_i represents a mean of an ith input parameter from the plurality of input parameters;

x_j represents a mean of a jth output parameter from the one or more output parameters;

σ

_i represents a standard deviation of the ith input parameter from the plurality of input parameters;

σ

_j represents a standard deviation of the jth output parameter from the one or more output parameters; and

|S_ij| represents sensitivity of the jth output parameter to the ith input parameter of the computational model.

13. The method according to claim 11, wherein generating further includes:

creating a neural network computational model;

training the neural network computational model using the data records; and

validating the neural network computation model using the data records.

14. The method according to claim 11, wherein determining further includes:

determining a candidate set of input parameters with a maximum zeta statistic using a genetic algorithm; and

determining the desired statistical distributions of the input parameters based on the candidate set,

wherein the zeta statistic ζ

is represented by;

$\mathrm{\zeta <br><br>}=\stackrel{j}{\underset{1}{\sum <br><br>}}<br><br>\stackrel{i}{\underset{1}{\sum <br><br>}}<br><br><br><br>S_{\mathrm{ij}}<br><br><br><br>\left(\frac{{\mathrm{\sigma <br><br>}}_i}{{\stackrel{\_}{x}}_i}\right)<br><br>\left(\frac{{\stackrel{\_}{x}}_j}{{\mathrm{\sigma <br><br>}}_j}\right),$

provided that x_i represents a mean of an ith input;

x_j represents a mean of a jth output;

σ

_i represents a standard deviation of the ith input;

σ

_j represents a standard deviation of the jth output; and

|S_ij| represents sensitivity of the jth output to the ith input of the computational model.