Hybrid uncertainty analysis of the nozzle guide vane surface temperature in hydrogen-fueled gas turbine

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 components in the gas turbines are subject to variabilities which can make a significant influence. Therefore, uncertainty analysis is essential. This paper investigates the surface temperature profile of a nozzle guide vane in a hydrogen-fueled gas turbine by considering both stochastic and interval uncertain parameters. The Polynomials Chaos Expansion (PCE)-Chebyshev method is applied to execute the hybrid uncertainty analysis. Coolant flow rate is considered as stochastic uncertain parameter and turbine inlet temperature profile is considered as interval uncertain parameter. A new reliability index is defined and compared with the conventional failure probability. The statistical moments of the interval bounds and the interval bounds of the statistical moments are estimated. The numerical results indicate that the results of PCE-Chebyshev method are in good agreement with the results of Scanning-Monte Carlo simulation method. The global bounds of vane surface temperature are assessed to examine the scope of the impact of uncertainties. Reliability analysis points out that under the influence of uncertainties, the locations with the highest safety on the vane are at the relative radial positions of 0.0 and 0.7, while the locations with the highest risk are at the relative radial positions of 0.4 and 1.0, which constitutes a foundation for the safety design of vane structures and provides a reference for subsequent robust optimization and refined design.