A combined experimental, numerical, and data consistency approach for the characterization of temperature distribution in a MILD combustion furnace

This work aimed to compare the non-intrusive Rayleigh scattering technique and intrusive suction pyrometry for temperature measurements in a semi-industrial MILD combustion furnace. The temperature trends obtained with the two measurement techniques were consistent. However, in the zone characterized by the jet development and the mixing process between the reactants and the recirculated burnt gases, laser-based measurements brought different temperature values, which compared favorably with model predictions. CFD simulations were used to shed light on the disturbances induced by the suction pyrometer and their impact on the fluid dynamics field. It was found that the suction pyrometer disrupts the flow inside the furnace, leading to an asymmetry of the system. In addition, the temperature measured using the suction probe is an average over the sucked volume which itself is variable in nature as, for a given suction volumetric flow rate, the sucked volume depends on the local mixture velocity and density. The present analysis can help reconcile inconsistencies found in CFD modeling in which probe sampling measurements were used for validation.