Flame Detection on Swirl Burner Using ECT With Dynamic Reconstruction Algorithm Based on the Split Bregman Iteration

To achieve increased combustion efficiency and reduced pollutant generation, methods to describe the spatial distribution of combustion parameters, such as combustion intensity and temperature are desired. Electric capacitance tomography (ECT) is by nature sensitive to the electrical properties of a flame, facilitating combustion intensity imaging. In this paper, a novel image reconstruction algorithm is developed. It is based on an enhanced regularization methodology and adequate fusion of prior knowledge in the optimization of the objective functional through which a higher spatial resolution and lower errors of the reconstructed images are realized. Simulation results show very good agreement of the reconstructed images with typical phantoms. Also, a series of experiments are carried out for a range of flow rates of fuel and air. Data are collected by the ECT sensor and hardware, and the corresponding images are reconstructed by the new algorithm. The experiments demonstrate the advantages of the new method and ECT images correspond well with the variation of the combustion intensity. Further analysis agrees well with the established principle of combustion.