BAYESIAN APPROACH FOR GAS CONCENTRATION RECONSTRUCTION BASED ON TUNABLE DIODE LASER ABSORPTION SPECTROSCOPY

摘要

Systems and methods construct a two-dimensional (2D) gas concentration map (image) from data from one or more laser-based absorption projection paths. Embodiments use a Bayesian approach to construct a 2D gas concentration map of the combustion zone cross-sectional plane by modeling the 2D map as a Gaussian process (GP). The GP models the global correlation between all the pixels in the 2D map. The data from one or more projection paths is propagated to all the pixels in the map instead of just propagating to the local pixels. The correlation between the map pixels is used to propagate the projection path data from the map pixels traversed by the projection path to the other map pixels.;