Abstract:

Some very precious professional test apparatuses are needed conventionally to determine the adiabatic temperature rise of concrete and the thermal diffusion coefficients of concrete surface. These apparatuses are very expensive and there may be errors in the test results due to different environmental conditions between laboratory and site. In view of this problem, based on the 3-D FEM of nonsteady temperature field and the genetic algorithm of inverse problem optimized solution, non-adiabatic temperature rise test of concrete cube on construction site is carried out to distinguish and ascertain concrete thermal parameters, including heat exchange coefficient, adiabatic temperature rise and law coefficient of adiabatic temperature rise. The results show that these parameters can indicate the real thermal properties of concrete in site. As one kind of optimization method, the genetic algorithm is simple and universal with good adaptability and high efficiency.

Key words: genetic algorithm, thermal parameters, back analysis, temperature field, adiabatic temperature rise