Leakage detection for hydraulic IGV system in gas turbine compressor with recursive ridge regression estimation

Abstract

The failure of the hydraulic Inlet guide vane (IGV) system needs to be avoided in the gas turbine compressor, since the IGV system is critical for the function and efficiency of the gas turbine and its fault can even cause the gas turbine to jump off the power grid. This paper investigates the detection of external and internal leakages, whose levels can be represented through corresponding leakage coefficients, at the cylinder in the hydraulic IGV system. Based on the dynamic model, we propose the recursive ridge regression parameter estimation method to detect and isolate different leakage coefficients under varying load. The developed algorithm is verified through experiments in a hydraulic IGV emulator. Based on experimental results, the proposed scheme can estimate leakage coefficients with good performance.