Assessment of Regularized Reconstruction of Three-Dimensional Temperature Distributions in Large-Scale Furnaces

Abstract

Three-dimensional (3-D) temperature distributions in large-scale furnaces can be reconstructed from temperature images captured by CCD cameras through inverse calculation of radiative heat transfer. This article assesses the reconstruction method for 3-D temperature distribution using the Tikhonov regularization by simulation. In a 10 m  10 m  20 m furnace with different optical thicknesses, the 3-D temperature distribution was reconstructed with errors in measurements and radiative properties of the combustion medium, and the influence of the radiative contribution from the wall surfaces. The results show that the Tikhonov regularization method is robust and, when the optical thickness is within the range from 1 to 15, the reconstruction errors of the 3-D temperature distribution in the furnace are less than 6%, which is acceptable in applications.

On the treatment of scattering for three-dimensional temperature distribution reconstruction accuracy in participating medium

“…2. The largest measurement error was checked for r = 0.1 [12] and the average divergence was also shown in Fig. 2.…”

Experimental investigation on simultaneous measurement of temperature distributions and radiative properties in an oil-fired tunnel furnace by radiation analysis

“…It should be mentioned that using the uniform radiative properties to reconstruct the temperature distribution in the inhomogeneous system may take some errors. This problem has been discussed and assessed in a 3-D furnace [27]. It has proved that the maximum reconstruction error of temperature is less than 6%, when the average values of radiative properties are used to replace the non-uniform distribution of radiative properties and measurement error is r = 0.1.…”

Recent progress in computational thermal radiative transfer

“…Zhou et al [87] proposed a flame temperature image-formation model, in which the two-color method is used to get a reference temperature, and a flame temperature image can be calculated from a monochromatic image in a color flame image, leading to the quantitative model relating, to the temperature measurement and the temperature distribution inside a furnace. Lou et al [89] examined the influence of optical thickness of industrial combustion systems on the reconstruction algorithm by the regularization, and the results showed that this algorithm can give acceptable results as optical thickness varies from 1.0 to 15.0. An improved Tikhonov regularization method was used to set up a reconstruction algorithm for the three-dimensional temperature distributions inside industrial furnaces [88] .…”