Stall Precursor Determination of an LM-2500 Gas Turbine

This thesis presents an analysis of data taken from several stall initiation events for an LM-2500 gas turbine engine. Specifically, the time series of three separate pressure signals located at stages 3, 6, and 15 were analyzed utilizing fast Fourier transform, power spectral density, and an autocorrelation technique to determine the best and most reliable indicator to stall. The spectral analyses performed showed that rotating precursor waves that travel at approximately half rotor speed were the best indicators. Several algorithms were used and it was determined that stall wave perturbations can be identified reliable about 880 revolutions prior to the stall. This work indicates that a single pressure signal located at stage 3 on an LM-2500 gas turbine can be used to give advance warning to a stall more than 2 seconds prior to the stall event.