ABSTRACT
Determining thermal diffusivity using flash diffusivity tests at high temperatures is investigated using parameter estimation. One aspect is the development of a method for determining two different heat transfer coefficients, one at the heated face and one at the opposite face. Both simulated exact and experimental data are used to illustrate the procedure. Although the heat transfer coefficients are different, assuming the heat transfer coefficients in the estimation process are the same does not significantly affect the estimates of the thermal diffusivity.
Insight into the estimation of thermal diffusivity and other parameters is obtained from a study of the sensitivity coefficients. Although the thermal diffusivity is the primary parameter of interest, a measured signal proportional to the temperature rise also depends on the heat transfer coefficients and energy input, which are called nuisance parameters (if they are not of interest). As the temperatures increase above 1500°C, the heat losses become very large and greatly influence the temperature response. By using insights from the study of the sensitivity coefficients for each of these parameters, the thermal diffusivity can be estimated despite the large heat losses.
