Application of Adaptive GPA to an Industrial Gas Turbine Using Field Data
Abstract The gas turbine inspection activities provided by the manufacturers and user maintenance scheme may be different from each other. To accommodate the difference, performing engine diagnostic as a condition-based monitoring technique is necessary to support Asset Performance Management (APM) adopted by the gas turbine users to improve the scheme. This paper provides an application of a novel Adaptive Gas Path Analysis (Adaptive GPA) to diagnose performance and health condition of a GE industrial gas turbine MS5001PA operated by PT Pupuk Kaltim (PKT). In the application, an engine thermodynamic model is constructed, adapted, and validated on the actual engine performance based on its gas path measurements. To estimate the health condition from the degraded engine data, two steps are applied in the Adaptive GPA diagnostic process. The first step is the estimation of degraded engine performance status and the second step is the prediction of engine health status at the gas turbine component level. Adaptive GPA results show that satisfactory predictions of the engine degradation have been achieved. In other words, the compressor has been predicted 5.56% degradation in flow capacity and 4.26% degradation in efficiency respectively, which is an indication of compressor fouling. Combining the diagnostic results, manufacturer’s recommendations, and user maintenance strategy, it is relatively safe and allowable to increase the maintenance inspection interval from 12,000 to 16,000 hours. Therefore, the adaptive GPA is proven to be beneficial to support condition-based maintenance decisions.