Development of stochastic finite prediction element model for CLT floors

One of the important challenges for wooden construction is to assure good sound insulation in low-frequency range. To that end, it is important to establish reliable prediction tools to reduce the number of experimental tests and the costs. The study reported here is a first investigation aiming at developing a reliable numerical model to predict the vibration behavior of a cross-laminated timber (CLT) floor in low-frequency range. More specifically, and in order to give more meaningful and useful inputs, stochastic simulations were applied to the CLT finite-element (FE) model, in using the measurement data as the canonical reference. A comparison between the conventional manual calibration and the stochastic calibration by tuning the material properties of the target modeling was subsequently made to evaluate the performance of the stochastic method employed. It was shown that this stochastic method gives an alternative to mimic the dynamic behavior of wooden construction elements without tedious manual calibrations when the data base of the material properties is missing. Meanwhile, the stochastic method included in the FE model also allows to take into account the uncertainties of the wooden materials which are not always under control because of the diversity and nature of wood.