Automated diagnostic system
First Claim
1. A diagnostic tool for troubleshooting a system based on status information, comprising:
- a diagnostic file containing lists of assemblies within the system, lists of test definitions for evaluating the status information, and numeric relationships between the listed assemblies and the listed test definitions;
a software program for evaluating the status information using the diagnostic file and providing a diagnostic report of the evaluation; and
a computer for running the software program and accessing the diagnostic file;
wherein the diagnostic file is structured so that the test definitions are each independent and the numeric relationships are each reversibly traceable.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention is directed to an automated diagnostic tool having a soft structure architecture that can be easily and quickly used to troubleshoot systems and subsystems, i.e., assemblies, on aircraft such as the AH-64 helicopter. The tool includes a laptop computer loaded with a troubleshooting software program, which evaluates assembly status messages traveling on a 1553 digital communications bus that links the aircraft systems and subsystems together. The tool obtains the messages by listening on the 1553 bus. The program includes listings of the assemblies, a set of pass/fail tests for evaluating the status information, and a set of relationships that numerically correlate test outcomes with different assemblies. Based on the test results and the relationships, the tool indicates which assemblies are the most likely cause of equipment malfunctions indicated by the test outcomes. The assemblies, tests, and the relationships between the tests and the assemblies are independent of each other, and the relationships are reversibly traceable.
100 Citations
27 Claims
-
1. A diagnostic tool for troubleshooting a system based on status information, comprising:
-
a diagnostic file containing lists of assemblies within the system, lists of test definitions for evaluating the status information, and numeric relationships between the listed assemblies and the listed test definitions;
a software program for evaluating the status information using the diagnostic file and providing a diagnostic report of the evaluation; and
a computer for running the software program and accessing the diagnostic file;
wherein the diagnostic file is structured so that the test definitions are each independent and the numeric relationships are each reversibly traceable. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A method for evaluating assemblies within a system that are connected via a bus, comprising the steps of:
-
listening on the bus for status information regarding the assemblies;
evaluating the status of the assemblies by performing the substeps of;
testing the status information by comparing predetermined components of the status information with test values;
correlating the test results with assemblies based on numeric values that indicate a strength of a causative relationship between each test result and an assembly;
assigning a probability for each of the assemblies based on the correlation;
selecting ones of the assemblies for inclusion in a diagnostic report;
presenting the diagnostic report, including the assigned probability for each selected assembly. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27)
receiving corrective feedback from an operator with respect to a probability associated with a selected assembly and listed in the diagnostic report; and
altering the numeric values for the selected assembly based on the corrective feedback.
-
-
18. The method of claim 17, wherein a magnitude of the alteration of the numeric values is determined by a case based adaption rate value.
-
19. The method of claim 11, wherein assemblies are selected based on a comparison of the probability assigned to each assembly with an acceptance level value.
-
20. The method of claim 19, wherein assemblies having an assigned probability that is greater than the acceptance level value are selected.
-
21. The method of claim 11, wherein the numeric values for each assembly include at least one of a positive likelihood and a negative likelihood.
-
22. The method of claim 11, wherein the probability for each assembly is based on a failure rate for the assembly.
-
23. The method of claim 11, further comprising the step of assigning an initial probability for each of the assemblies equal to a value of 0.5.
-
24. The method of claim 11, wherein the substeps of correlating and assigning comprise the steps of:
-
calculating the probability for each of the assemblies with respect to one of the test results using a first equation Pi+1=(Pi*Ppos)/[(Pi*Ppos)+(1−
Pi)*Pneg]; and
assigning the probability Pi+1 to the assembly;
whereinPi is the probability previously assigned to the assembly, Ppos is a positive likelihood indicating the strength of the causative relationship between the test result and the assembly, and Pneg is a negative likelihood indicating the strength of the causative relationship between the test result and the assembly.
-
-
25. The method of claim 24, wherein Pneg=(1−
- Ppos).
-
26. The method of claim 24, wherein the substeps of correlating and assigning further comprise the steps of:
-
determining which assembly has the highest assigned probability with respect to the one test result;
calculating for the determined assembly a probability for each test result using the first equation;
selecting the test result corresponding to the highest calculated probability for the determined assembly;
performing the step of calculating the probability for each of the assemblies with respect to one of the test results, using the selected test result as the one test result; and
repeating the steps of calculating the probability for each of the assemblies, assigning the probability Pi+1 to the assembly, determining which assembly has the highest assigned probability, calculating for the determined assembly a probability for each test result, selecting the test result corresponding to the highest calculated probability, and performing the step of calculating the probability for each of the assemblies using the selected test result, until at least one of the probabilities rises above a threshold.
-
-
27. Method of claim 26, wherein the substep of selecting ones of the assemblies for inclusion in the diagnostic report includes the step of selecting each assembly corresponding to each at least one probability above the threshold.
Specification