Gas turbine diagnosis using a fault isolation enhanced GPA

Abstract

Gas Path Analysis (GPA) and its different derivatives have been developed for more than thirty years and used widely and successfully by many gas turbine manufacturers and operators. In gas turbine gas path component diagnosis, it has been recognized for a long time that GPA would be more successful if degraded components could be located. Unfortunately, only the deviation of measurable parameters is monitored in operation and information about the degraded components is normally not available. In this research, a two-step diagnostic approach is introduced, where a pattern matching method is used first and further developed to isolate degraded components; then Gas Path Analysis is applied to assess the quantity of degradation. A gas turbine performance simulation program, Cranfield University TURBOMATCH, has been modified to simulate the diagnostic process. A model gas turbine engine similar to Rolls-Royce aero AVON is used to test the effectiveness of the approach. It is found that the developed fault isolation method can isolate degraded components accurately and enhance the effectiveness of the quantitative assessment of the degradation with Gas Path Analysis (GPA) in gas turbine diagnostics. Copyright © 2004 by ASME