Diagnostic trend analysis

• US 5,951,611 A
• Filed: 11/18/1996
• Issued: 09/14/1999
• Est. Priority Date: 11/18/1996
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: La Pierre, Lee R.
• Primary Examiner(s): Cuchlinski, Jr., Willliam A.
• Assistant Examiner(s): Hernandez, Olga
• Application Number:
  US08/748,745
• Time in Patent Office: 1,030 Days
• Field of Search: 701/29, 364/576
• US Class Current: 701/29
• CPC Class Codes: G05B 23/0232 based on qualitative trend analysis, e.g. system evolution
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA

Abstract

A diagnostic trend analysis method is used to analyze engine trend data for aircraft engines. Appropriate trend parameters are calculated from raw trend data and wild points associated with the calculated trend parameters are eliminated. Abnormal shifts and changes are detected at the most recent point for each trend parameter. The abnormal shifts and changes in each trend parameter are then evaluated. The evaluation is used to filter out spurious symptoms and provide accurate maintenance recommendations for real engine trend shifts.

Claims

1. A diagnostic trend analysis method for analyzing engine trend data for aircraft engines, the method comprising the steps of:

calculating appropriate trend parameters from raw trend data;

eliminating wild points associated with the calculated trend parameters;

detecting abnormal shifts and changes at a most recent point for each trend parameter and differentiating between a real change in engine performance and a faulty measurement or data scatter;

evaluating the abnormal shifts and changes in each trend parameter; and

using the evaluation to filter out spurious data and provide accurate maintenance recommendations for real engine trend shifts.

2. A diagnostic trend analysis method as claimed in claim 1 further comprising the step of calculating corrected parameters to determine actual trend parameters.

3. A diagnostic trend analysis method as claimed in claim 1 further comprising the step of generating a fault template indicative of abnormal shifts and corresponding maintenance recommendations.

4. A diagnostic trend analysis method as claimed in claim 3 further comprising the step of comparing a set of actual trend symptoms to the fault template to identify a possible cause for each of the actual trend symptoms.

5. A diagnostic trend analysis method as claimed in claim 4 further comprising the step of evaluating the match between the set of actual trend symptoms and the fault template.

6. A diagnostic trend analysis method as claimed in claim 5 wherein the step of evaluating the match further comprises the step of scoring proximity of actual trend parameter shifts to each set of shifts in the fault template.