Three-dimensional reconstruction of flame based on computed tomography of chemiluminescence

High-temperature field detection has important application value in the industrial and military fields. The three-dimensional structure reconstruction of flame is of great significance to carry out the finite element division of the temperature field. However, because the flame is a self-luminous fluid, the flame pictures obtained directly by the camera lack characteristic information, and the use of many three-dimensional reconstruction technologies directly performs topographic reconstruction with low accuracy. In this study, MATLAB software was used to conduct simulation experiments to study the accuracy of the 3D structure of flame reconstructed by chemiluminescence computed tomography. A single Gaussian peak was used to simulate a symmetrical flame, and a 2D projection of the flame was obtained according to a simplified chemiluminescence tomography model. The additive algebraic reconstruction method was used in this study. (Additive Algebra Reconstruction Technique) reconstructs the three-dimensional structure of the flame. The resolutions of 256px, 512px, 1024px triple-projection images are set respectively, as well as the different number of channels. The mean error is used for quantification to evaluate the accuracy of the reconstruction results. The experimental results show that the resolution of 512px and the number of channels of 13 converge after 8 iterations, and the reconstruction accuracy is better than that of the projection resolution of 256px. At the same time, the reconstruction speed of the ART algorithm at 512px is much faster than that at 1024px. At the same resolution, the reconstruction accuracy increases with more channels.