A disc to air heat flux error and uncertainty analysis applied to a turbomachinery test rig design

Abstract

This article describes a Monte Carlo simulation-based error and an uncertainty analysis for values of disc to air heat fluxes as part of the design of an experimental axial turbine test rig. This work is of interest for those who study heat transfer and measurement or the design and use of experimental test rigs. An inverse analysis of theoretical disc surface temperatures was performed for different thermocouple configurations to compare the errors and uncertainties resulting from each to establish whether there was any configuration that would return the lowest magnitudes of error and uncertainty and hence influence the location of the proposed instrumentation. It is shown that great care needs to be taken when using an analysis of this kind together with temperature measurements having realistic and typical uncertainty values. This is because such an analysis is purely analytical, and any small fluctuations in the inputs, such as typical thermocouple uncertainties and noise, result in the process of an inverse analysis becoming unstable. This instability has two effects: (a) the returned values of heat flux have an inbuilt bias error and (b) the magnitudes of uncertainty can exceed>100 per cent.