Search

New materials and structures are applied in order to maintain the engineering products more and more efficient. Different types of thermal analysis/modelling of such materials and structures are also needed for the proper planning or design. The thermal analysis/modelling processes are supported by many commercial codes but in some cases, the experts might have applied a self-developed method also. The success of the analysis/modelling depends on many factors, but the basis is the correct application of the heat conduction theory which is more important in case of transient thermal processes. As the used tools are more and more powerful, the users pay less attention to the theory background, according to the author experience. It seems that the long tradition of the inconsistent application of the temperature-dependent thermal diffusivity is continuing, even strengthened by the increased availability of data of the temperature-dependent thermal properties. Regarding the origination of the thermal diffusivity, this might lead to the misinterpretation or misunderstanding of the nonlinear heat conduction problem solution procedures. This paper calls attention to avoid this, shows what kind of mistakes might occur related to this widely observable practice, summarizes the proper application and interprets where and how one could find room for the temperature-dependent thermal diffusivity. [ABSTRACT FROM AUTHOR]

Simultaneous inverse identification of the temperature-dependent volumetric heat capacity and thermal conductivity of a solid material based on transient temperature histories was studied. The inverse problem was defined according to the evaluation of the BICOND thermophysical property measurement method. The material property functions were defined by several data points with linear interpolation between them. In this way, no previous information is needed about the expected temperature dependency of the material property functions. The inverse problem was solved by a real-valued genetic algorithm using simulated measurement results. New genetic operators (smooth initialization and smooth mutation) were developed and applied. The accuracy of the inverse solution was studied in two test cases including linear, square, and sinusoidal functions. The effects of regularization and random noise in the temperature histories were also analyzed. Based on the results, the proposed method is likely to be effective in evaluation of real measured temperature histories to simultaneously determine temperature-dependent thermophysical properties. [ABSTRACT FROM AUTHOR]