Interval uncertainty analysis for the nozzle guide vane surface temperature profile in hydrogen-fueled gas turbine

Abstract

Hydrogen has the advantages of zero-carbon emission, high heating value per mass, and wide distribution. Hydrogen-fueled gas turbines represent a highly promising way to take advantage of hydrogen energy. The ubiquitous uncertainties can cause significant variations in the performance of turbomachinery. In this paper, the uncertainty quantification of a turbine nozzle guide vane in a hydrogen-fueled gas turbine is conducted to investigate the effects of uncertainties on the vane surface temperature profile. The peak temperature on the vane surface and its location have been specified by the analytic cooling model. Due to the difficulty of obtaining the probability density function for uncertain parameters, the Chebyshev interval method and the Legendre interval method are used. The coolant flow rate and turbine inlet temperature radial profile are two uncertain-but-bounded parameters. A new sensitivity index is defined to conduct global sensitivity analysis and the satisfaction degree of interval is employed to measure the reliability of the guide vane. It has been observed that the uncertainty in the inlet temperature profile leads to a higher satisfaction degree, thus having a more negative impact on the reliability of the vane. Sensitivity analysis points out that the significance of coolant flow rate and inlet temperature profile in terms of the uncertainty in the vane peak temperature is essentially the same, contributing to about 50%. The research reveals that the Chebyshev method has smaller errors.