Temperature and Gas Concentration Imaging of LAS Tomography from Gaussian Filtered SART with Artifact Removal

Laser absorption spectroscopy (LAS) is widely applied for tomographic imaging of temperature and gas concentration in combustion fields. To improve the imaging accuracy and reduce artifacts, the Gaussian filter is added in the iteration of simultaneous algebraic reconstruction technique (SART) as a smoothing operator, and an artifact removal method is developed based on the correlation between flame temperature and concentration. Symmetrical and asymmetrical distributions of temperature and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{H}_{2}\mathrm{O}$</tex> mole fraction obtained from the computational fluid dynamics (CFD) simulation were reconstructed in the numerical verification. The proposed Gaussian filtered SART with artifact removal performed well with the fewest artifacts and lowest reconstruction errors compared to typical SART-based methods. The exit of a high-temperature wind tunnel was measured for experiment verification, and temperatures at the center of the wind tunnel exit in 11 different cases were measured by the thermocouple as a reference. Temperature and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{H}_{2}\mathbf{O}$</tex> mole fraction distributions reconstructed by the proposed method were smooth, and its variation trend was consistent with the real condition. The maximum relative error of temperature of the proposed method was only 5.29%, much smaller than that of the typical SART with a median filter. Both numerical and experiment verifications demonstrated the ability of the proposed method in high-accuracy tomographic imaging.