부산대학교 도서관

Numerous power generation technologies rely on plate geometry for the purpose of transporting heat through and across mediums. In the case of nuclear technology, this is relevant to that of plate-type-fuel and compact heat-exchangers; these components have high surface area to volume ratio provide ideal capacity for heat transfer to and through their respective parts. However, under highly advective conditions these plate-type geometries are susceptible to buckling and vibration which have possible safety-related implications associated with them. These complex phenomena occurring through fluid-structure-interactions have traditionally been quite difficult to predict using computational tools given their stiffly coupled nature. While recent efforts have demonstrated well aligned agreement with the use of advanced computational tools to predict flutter and buckling of unique mechanical geometries these tools have been found to consume significant computational resources not available to many within the community. This study investigates an industry tool's ability to compute the natural frequency of plates under various boundary conditions for the purpose of demonstrating relevance and limitations of its computing capacity when compared against experimental data to develop an objective basis for the use of such a tool in design and safety related predictions of components which comprise plate-type geometries. The goal of this study is to demonstrate a new method using an off-the-shelf software tool which is capable of operating on a standard desktop workstation and can readily predict the dynamic response of a mechanical structure in a fluid environment – a capability not previously demonstrated or otherwise shown to be successful in literature.A set of experimental plates were characterized to understand their dynamic response in both air and water. In an attempt to gain further insight into the vibration of the experimental test plates, a numeric benchmark study was performed to calculate their fundamental frequencies in both air and water, employing a novel modeling method to simulate submersion; one which allowed both the plate and the fluid domain to be modeled exclusively in a computational structural mechanics software package. Results show that the air simulation compares well with the analytic solutions; however comparison against experimental data is challenging, with no obvious trends present. Given the available experimental data, select air simulations are quite accurate, and careful selection of numeric boundary conditions yields representative results for alternative experimental boundary conditions. For simulations of plates submerged underwater, larger data sets for both experimental and numeric will be required to establish trends with a high degree of confidence; however preliminary results suggest the method can be accurate to a first order of magnitude approximation, requiring minimal computational resources.