Analysis of the Impact of Degradation on Gas Turbine Performance Using the Support Vector Machine (SVM) Method

Degradation is an important aspect in the operation and maintenance of gas turbines since it affects maintenance costs substantially. Hence, the study of degradation in terms of recoverable and non-recoverable degradation is crucial to formulate a correct maintenance strategy and, as a result, achieve optimum maintenance cost. In this paper, the impact of recoverable and non-recoverable degradation towards compressor discharge pressure, fuel gas flow, and exhaust gas temperature are measured during the start of run period that reflects the time period from the new gas turbine condition to the first scheduled offline crank wash, which normally approximates to 8000 running hours. For the study, a three-unit single speed light industrial gas turbine that drives an electrical generator to power up an offshore platform located in a tropical climate is considered. The measurement of the parameters has been conducted using the support vector machine (SVM) method.