Condition-based, diagnostic gas path reasoning for gas turbine engines

This paper addresses the requirements and concept for a condition-based, diagnostic gas path reasoner algorithm for the aggregation, classification and prediction of gas path faults/anomalies in gas turbine engines. A more accurate maintenance recommendation is made based upon all available information about the state of the engine. We propose a scheme that gathers the results of different diagnostic methods and leverages the advantages of each one. Such a fusion scheme holds the promise to deliver a result that is at least as good as the best of the algorithms used, and potentially much better because of the leveraging of complementary information of the other diagnostic algorithms. There are many issues involved with gas path reasoning. We present an approach that tackles the problems and focuses in particular on the system issues involved in implementing an advanced, condition-based, diagnostic gas path reasoner.