Infrared Species Limited Data Tomography using Kalman Filtering

Abstract

In infrared species limited data tomography the spatial concentration distribution of an unknown species, for example the air/fuel ratio in a combustor, is inferred from the attenuation of multiple collimated light beams shone through the measurement field. The resulting set of linear equations is rank-deficient so it is necessary to introduce prior assumptions about the smoothness and nonnegativity of the distribution to recover a solution. This paper investigates whether the Kalman filter can be used to incorporate additional information from the observed time-evolution of the attenuation measurements to further improve the reconstruction accuracy, but finds that performing a series of static measurements is both more accurate and computationally efficient.