A flame picture is worth a thousand words

Dr. Victor Chernov

Combustion

My research on sooting flames centers on an innovative image reconstruction algorithm that predicts how a flame would appear to a digital camera. This has two key applications: First, it provides a simplified and more reliable method for validating sooting flame simulations by directly comparing synthetic and real camera images. Second, it offers the potential for a novel measurement tool by iteratively adjusting simulation parameters until the synthetic image matches the captured image, potentially using machine learning for optimization. Future work will extend this approach to multi-dimensional flame reconstruction, including non-axisymmetric and turbulent flames, and incorporate multiple camera views.