Modelling of the Performance of Temperature Tomography systems with IR Laser Sources

Michael Wood, Edward Cheadle, Peter Ireland†, John Black†, Hugh McCann, Krikor Ozanyan School of Electrical and Electronic Engineering,

The University of Manchester, Manchester M13 9PL, United Kingdom

†Rolls­Royce plc, Derby DE4 8BJ, United Kingdom

ABSTRACT

In industrial processes involving combustion, information about the spatial distribution of the temperature and concentration fields may be used in the design of these systems to achieve higher operating efficiencies; for example, by identifying and removing hotspots the maximum temperature the components are exposed to can be reduced, alleviating the cooling requirements. Industrial combustion subjects can impose limits on both the longevity and efficiency of the measurement system to be deployed because of the high temperatures, challenging the tolerances of the materials from which components are made. Furthermore, access restrictions have to be taken into account, which makes the design and implementation very sensitive to geometry. In this work we seek a methodology of obtaining a spatially resolved temperature field of the gases within combustion environments using Tomography, because of its non­intrusive character requiring only periphery access.

We show that the values of the local absorption coefficients of combustion gases directly depend on the temperature; however it is not possible to express the measurements in terms of line integrals of temperature. Consequently, we introduce an alternative method: from the near IR transmission measurements at different frequencies we reconstruct spectrally resolved images of the absorption coefficient in the measurement region, the values of which at each point can then be used to calculate the temperature at that point, thus generating a temperature map. The local absorption coefficient is predicted at each spatial point as a function of the frequency of the incident light, using the HITRAN database in the region of the IR overtones of the molecular fingerprints. The geometry­dependent line integrals are then calculated, to simulate the transmission measurements across the projection beams for each frequency.

Keywords Temperature Tomography, line integrals, HITRAN database, infrared spectroscopy