Flame Light Field Tomography in Combustion Chambers of Maximum a Posteriori Estimate using Bayesian Prior

Accurate temperature distributions are crucial in ensuring reliable engine operations, studying turbulent combustion, and optimizing combustion chambers. Non-invasive light field (LF) imaging provides a means to instantly measure three-dimensional (3D) data by microlens arrays. However, visualizing confined combustion brings challenges such as occlusions, interferences from walls, radiative properties effects on data, and limited projections. To overcome these challenges, we develop an accurate model for confined combustion flame using LF imaging and radiation distribution factor-based backward Monte Carlo (RDF-BMC), decoupling radiative properties and temperature. To address ill-posed inversions and fluctuations in reconstructions, we propose a Maximum a posteriori (MAP) estimation with prior smoothing based on Bayesian theory. Simulation studies verify the feasibility of the proposed method in a confined combustion field and reconstructed temperature under different noise levels. The proposed model and methodology enable instantaneous 3D flame diagnostics, providing an innovative tool for combustion mechanism analysis and optimal combustor design.