Performance-based prognosis scheme for industrial gas turbines

In this paper, we present a novel method for performance-based prognostics of industrial gas turbines. The concept of performance adaptation is implemented through a dynamic engine model that is developed in Matlab/Simulink environment to diagnose the health of the gas turbine. The proposed method is tested under variable operating conditions at both steady state and transient operational modes for estimating and predicting the compressor degradation. Different types of mathematical representations are used to fit the diagnosis results and consequently prognose the performance behavior of the engine. The results demonstrate the promising prospect of our proposed method for predicting accurately and efficiently the performance of gas turbine compressors as they degrade over time. © 2015 IEEE.