Data-Driven Model Selection Study for Long-Term Performance Deterioration of Gas Turbines

Abstract

Performance of gas turbine engine (GTE) deteriorates with structural aging. The availability of operating data from GTE and capability to perform data analysis, provides an opportunity to identify long-term performance deterioration and relate to more difficult to detect structural degradation. In this work, performance analysis of a low power rating and partially loaded industrial GTE was carried out by using a model-free data analytic approach. A performance index (ratio of power generation to fuel consumption) is proposed as the metrics for monitoring the engine performance, and monitor the long-term degradation symptom. A comparative model selection study has been conducted among three multivariable models to select the best model describing long-term performance deterioration of the GTE.