A generic gas turbine simulation tool has been used for the development of a gas path analysis model for the PW100 engine. The suggested method estimates the performance parameters of an individual engine model based on overhaul test cell acceptance data of a well-serviced engine. The method makes simultaneous use of data from readings collected at three different power levels, by selecting a preliminary design model, which gives optimal off-design agreement with the recorded data. Measured thermodynamic data has been augmented with the use of low pressure and power turbine areas to maximise the use of available information. Nine performance parameters are estimated from, in total, 32 thermodynamic and geometric data recorded at the three different power levels. The resulting model is subsequently used for fault detection for the same engine. The method was tested on two data sets, representing a well serviced and a faulty condition. The known low-pressure turbine nozzle fault was successfully detected based on the output from the gas path analysis tool. The analysis was repeated for the three single power settings. Analysis based on all three operating points produced a marginal improvement in diagnostic capability, compared to the best single point case and substantially better than the worst single point case.

This content is only available via PDF.
You do not currently have access to this content.