Temperature Calculation in an Uncooled Low-pressure Stage of a Heavy-duty Gas Turbine Using Conjugate Heat Transfer Analysis

抄録

In this paper, conjugate heat transfer (CHT) simulation is performed for a low-pressure stage of MGT-70, a heavy-duty gas turbine (GT) manufactured by MAPNA Group. Although the vane and the blade are uncooled, CHT analysis is performed to assess the validity of using the fluid temperature of an adiabatic simulation as the uncooled vanes or blades temperature, and also to model the heat transfer between root or shroud and vane or blade profile. To compare the resultant temperatures of CHT and adiabatic flow analysis both simulations are done, using the same boundary conditions. The vane shroud extends over the blade tip, which is of free-standing type, and there is not a shroud segment. In fact, the vane and blade share the shroud. In order to predict the shroud temperature more accurately, the vane and the blade are simulated simultaneously as a stage using appropriate interfaces. A single vane CHT simulation is also performed to evaluate the effect of blade tip flow on the shroud temperature. Furthermore, the cavity above the shroud, containing the cooling and sealing flow, is also included in the model to better prediction of the shroud temperature. In addition, the rim cavity and the labyrinth seal under the vane platform are included in the model to better predict the vane platform temperature and to capture the effect of purge flows on vane and blade temperature. Simulation results show that, although, the average bulk temperature of the profile in both CHT and adiabatic simulations are close to each other, there are great differences in temperature distribution over the suction side and pressure side. These differences are because of heat flux through the profile in CHT simulation, which results in a more realistic metal temperature distribution. Comparing the results of the single vane simulation and stage simulation no remarkable difference is observed in the temperature distribution, except for the shroud region above blade tip. This reveals that, although, the tip leakage flow is better captured in the stage simulation, it is only useful when the shroud temperature is of interest and it does not affect the vane profile temperature distribution. Finally, the inclusion rim cavity and labyrinth seal in the simulation helps to predict the mass flow distribution of purge flows and the effect of these flows on platform temperature distribution in vane and blade.