چكيده به لاتين
In the first step, this thesis considers stochastic model updating in rotating systems. Stochastic model updating methods consider model response variability and allocate them to the model parameters; however, they are not commonly employed in rotor dynamics, and deterministic approaches are still prevalent in this field. Due to the cost and efforts needed to set up experiments and to obtain outcomes that reflect actual characteristics of the machine, stochastic updating practices of industrial rotating systems are rarely reported in the literature. This thesis adopts an appropriate parameter selection procedure and suitable sampling strategy for stochastic model updating to investigate variability in the dynamic behavior of a complex turbo compressor rotor-bearing-support system, leading to successful parameter identification results. The compressor rotor is mounted on hydrodynamic journal bearings with speed-dependent stiffness and damping. Due to the rotating system complex model, a variance-based global sensitivity method is employed for parameter selection to eliminate non-influential parameters in the model updating and to alleviate updating complexity and computational burden. The Bayesian approach in the stochastic model updating is applied to estimate parameter uncertainty in the rotor with speed-dependent characteristics. Advanced Markov chain Monte Carlo sampling method using delayed rejection adaptive Metropolis (DRAM) algorithm is employed in the stochastic model updating. The updating procedure obtains marginal posterior probabilities of parameters, and uncertain parameter distributions are evaluated using maximum entropy criterion. In the second step, this thesis investigates the stochastic behavior of frictional supports that are examples of mechanical contact exhibiting nonlinear effects. Modeling uncertainty in the structures with mechanical contacts in nonlinear regime is a forward step in the design of engineering structures which is rarely reported in the references. Stochastic nonlinear models must be developed for such unknown structures. In this thesis, a stochastic identification algorithm is presented for updating of frictional contacts with nonlinear behavior, which has had successful results. In this algorithm, Bayesian method with two steps of identification using the likelihood function based on minimizing the Mahalanobis distance by considering the statistical characteristics of the test samples and the model, provides a reliable framework for updating the mean values and standard deviation of the parameters. It is indicate that Bayesian inference combined with the effective DRAM sampling method provides more accurate results with two-step updating than single-step identification in updating of parameter distribution. The proposed stochastic updating method was first implemented on a set of three degrees of freedom mass and spring with nonlinear contact and then on a standard test setup with a nonlinear support in the modal laboratory of the IUST. Elastoplastic Valanis model is used to estimate the nonlinear behavior of the contact surface. Measured responses of the structure were employed to identify the nonlinear restoring forces in the frictional support using force state mapping at several sessions of excitation. Elasto-plastic Valanis model is employed in order to estimate the nonlinear behavior of the contact interface in the structure. The parameters of Valanis model are considered as stochastic variables to account for measurement variability during repeated assembly and disassembly of structure. The statistical properties (mean values and standard deviations) of Valanis model parameters are identified using proposed approach based on Bayesian method applied to the restoring force in the contact interface. Finally, simulated model results are compared with experimental benchmark structure test data. The convergence between the statistical behavior of the predicted response and the measured data shows the efficiency of the proposed model in updating the nonlinear contact model.
