Algorithm for identifying tests to perform for fault isolation

• US 5,195,095 A
• Filed: 12/28/1990
• Issued: 03/16/1993
• Est. Priority Date: 12/28/1990
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Shah, Rasiklal P.
• Primary Examiner(s): ATKINSON, CHARLES
• Application Number:
  US07/635,442
• Time in Patent Office: 809 Days
• Field of Search: 371/15.1, 371/29.1, 371/22.1, 364/274.2, 364/276.3, 395/916, 395/575
• US Class Current: 714/26
• CPC Class Codes: G01R 31/2846 using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
• CPC Class Codes: G06F 11/2273 Test methods
• CPC Class Codes: Y10S 706/916 Electronic or computer, internal or network, circuit

Abstract

The present invention, in one embodiment, is a method for identifying a test to perform for fault isolation. The method comprises the steps of identifying a loop/string of components having at least one component therein which is suspected of faulty operation, identifying a test, from a listing of tests, which satisfies pre-determined conditions, and selecting the identified test for performance.

Claims

1. A method for identifying a test to perform for fault isolation, said method comprising the steps of:

(a) creating a diagram composed of blocks representing components of the system;

(b) representing connectivity of the components on the diagram;

(c) selecting a first start component;

(d) assigning a first label to the start component;

(e) identifying respective children components of the start component;

(f) assigning the respective children components of the start component respective labels, the respective labels being unique for each component and representing the component connectivity;

(g) selecting another start component if any component connectivity represented in the diagram is not represented by a label subsequent to performance of step (f) and, after selecting a second start component returning to step (e);

(h) storing the assigned component labels;

(i) upon generation of an error code, determining the component label of the component which generated the error code as a loop/string component;

(j) identifying upstream components of the loop/string component by sequentially deleting the right-most alphanumeric digit of the loop/string component label thereby generating a match label, and matching the match label with the labels of other components, any component label which matches the match label being tagged as an upstream component; and

(k) identifying downstream components of the loop/string component by comparing the left-most alphanumeric digits of labels of other components with the alphanumeric digits of the loop/string component label, any component label whose left-most alphanumeric digits match with the alphanumeric digits of the loop/string component label being tagged as a downstream component;

(l) identifying a test from a listing of tests, which satisfies predetermined conditions, and

(m) selecting the identified test for performance.

2. A method in accordance with claim 1 wherein the first start component is assigned the label "1" and a variable i is initially equal to "1", and each subsequent start component is sequentially assigned the label (i+1).

3. A method in accordance with claim 1 wherein the children components are sequentially assigned an alphanumeric digit starting with "1" and including the parent component label.

4. A method in accordance with claim 1 wherein the diagram is created on a graphical editor.

5. A method in accordance with claim 1 wherein the assigned component labels are stored within a central computer memory.

6. A method in accordance with claim 1 wherein a component is represented in the diagram as an independent component only if the component is capable of generating an error code.

7. A method in accordance with claim 1 wherein a component is represented in the diagram as an independent component only if the component has a probability of failure above a pre-selected failure probability.

8. A method for operating a computer for identifying a test to perform for fault isolation, comprising the steps of:

creating a diagram composed of blocks representing components of the system;

representing connectivity of the components on the diagram;

selecting a first start component;

assigning a first label to the start component;

identifying respective children components of the start component;

assigning the respective children components of the start component respective labels, the respective labels being unique for each component and representing the component connectivity;

selecting another start component if any component connectivity represented in the diagram is not represented by a label;

storing the assigned component labels;

upon generation of an error code, determining the component label of the component which generated the error code as a loop/string component;

identifying upstream components of a first loop/string component by sequentially deleting the right-most alphanumeric digit of the first loop/string component label thereby generating a match label, and matching the match label with the labels of other components, any component label which matches the match label being tagged as an upstream component; and

identifying downstream components of the first loop/string component by comparing the left-most alphanumeric digits of labels of other components with the alphanumeric digits of the first loop/string component label, any component label whose left-most alphanumeric digits match with the alphanumeric digits of the first loop/string component label being tagged as a downstream component;

identifying a test, from a listing of tests, which satisfies pre-determined conditions; and

selecting the identified test for performance.

9. A method in accordance with claim 8 wherein the first start component is assigned the label "1" and a variable i is initially equal to "1", and each subsequent start component is sequentially assigned the label (i+1).

10. A method in accordance with claim 8 wherein the children components are sequentially assigned an alphanumeric digit starting with "1" and including the parent component label.

11. A method in accordance with claim 8 wherein the diagram is created on a graphical editor.

12. A method in accordance with claim 8 wherein the assigned component labels are stored within a central computer memory.

13. A method in accordance with claim 8 wherein a component is represented in the diagram as an independent component only if the component is capable of generating an error code.

14. A method in accordance with claim 8 wherein a component is represented in the diagram as an independent component only if the component has a probability of failure above a pre-selected failure probability.