Identification of temperature-dependent thermal-structural properties via finite element model updating and selection

The objective of this study is to develop a strategy to identify the temperature-dependent properties of a thermo-elastic structure in an unsteady temperature environment, where time-varying material properties and thermal stresses are taken into account. The identification problem is formulated as an updating procedure of the finite element model. Due to the unsteady temperature environment, this procedure is based on a timevariant finite element model because the system matrices change over time. The temperature-dependent properties are expressed as low-order polynomials first. Then, an integrated objective function is established by using errors of the instantaneous frequencies and the sum of the highest order of the polynomials for all the parameters. Subsequently, the particle swarm optimisation is performed to minimise the above objective function to simultaneously determine the coefficient and the order of the polynomials. To demonstrate the effectiveness of the proposed procedure, the identification of a simply supported beam with an axially movable boundary subjected to an unsteady, uniformly distributed temperature field is presented. The numerical verification shows that the identified temperature-dependent properties well track the trends of the true values with high accuracy.