Tadashi Iijima

Modeling method of element Rayleigh damping for the seismic analysis of a 3D FEM model with multiple damping properties

Damping modeling is important for the accurate evaluation of the seismic response of structures. Our group previously reported a damping modeling method using element Rayleigh damping and evaluated the effectiveness using a simple lumped-mass model with multiple damping properties; however, the effectiveness of the method was not evaluated for three-dimensional (3D) finite element method (FEM) models with multiple damping properties. Moreover, further studies showed that the method needed to be improved to be applied to 3D FEM models. Therefore, the method has been improved to enable application to the seismic analysis of 3D FEM models, and the effectiveness of the method has been evaluated. The proposed method uses a weighted least-squares method to automatically determine the coefficients of element Rayleigh damping. The weighted least-squares method minimizes the differences between the modal damping ratios to be modeled and those given by element Rayleigh damping. Although all modal damping ratios in a simple lumped-mass model were used for damping modeling in our previous study, obtaining them for 3D FEM models is impractical because these models have more natural modes than simple lumped-mass models. Therefore, we used modal damping ratios below a cut-off frequency. The effectiveness of the proposed method was evaluated by comparing it with conventional methods in terms of the modeling errors related to the modal damping ratios and the maximum absolute acceleration. The proposed method tended to have lower errors than the conventional methods and is concluded to be more effective for the seismic analysis of 3D FEM models with multiple damping properties. The proposed method can automatically determine the coefficients of element Rayleigh damping and can more accurately model the damping properties of analysis models, indicating that the proposed method is a powerful tool for the seismic analysis of 3D FEM models with multiple damping properties.

Full Scale Simulation-Based Study on Dynamic Response of BWR Fuel Assemblies Under Seismic Loading

The purpose of this study is to investigate dynamic response behaviors of fuel assemblies in a boiling water reactor (BWR) under seismic loading. The core of BWR consists of several hundreds of fuel assemblies. They are supported with both top guide and fuel support and are surrounded by coolant water. It is important to grasp their dynamic response behaviors under seismic loading for securing the structural integrity of the fuel assembly itself as well as for assessing control rod scrammability. In this study, we employ two different numerical simulation methods of acoustic fluid-structure interaction (AFSI) developed by the present authors independently. The one is a three-dimensional parallel finite element method for AFSI problems with solid elements based on a partitioned coupling approach, while the other is a finite element method of beam elements for fuel assemblies combining added mass matrix, which represents coupled inertia effects caused by coolant water.Both methods are first applied to a problem of 36 fuel assemblies for numerical verification, and then applied to a problem of 368 fuel assemblies for validation. The latter problem was set up based on the demonstration test performed by the NUPEC (Nuclear Power Engineering Corporation) in 1986. Both simulation results agreed well with each other in all cases, and the simulated results also agreed well with the experimental ones. In addition, we have precisely discussed seismic response behaviors of the fuel assemblies, which were not shown in the demonstration test. Accordingly, we conclude that the both developed simulation methods are powerful tools to grasp the precise behavior of fuel assemblies of BWR under seismic loading and to improve the seismic safety design of BWR core.Copyright

A Simplified Analysis Method for Elasto-Plastic Seismic Response Using the Energy Balance Equation of a Maximum Hysteresis Loop

With increasing magnitude of design earthquake ground motions, it is necessary to develop methods of evaluating the seismic safety margin that are more exact than the current methods. However, a standard nonlinear analysis method requires step by step calculations of the numerical time integration scheme to obtain the seismic response. The authors present a new simplified analysis method of elasto-plastic seismic response. The proposed method is formulated by the energy balance between the input energy and the dissipated energy of an equivalent single degree of freedom model for actual equipment. Assuming the harmonic resonance of the single degree of freedom model, the maximum displacement response can be estimated conservatively. To verify the proposed method, static tests and vibration tests with cantilever-type specimens were performed. The vibration tests were conducted with sine, sweep down sine and random waves to verify the conservativeness of the proposed method. Comparisons of the maximum displacement between the tests and the proposed method show the conservative estimation of the displacement by the proposed method.Copyright

Seismic Margin Evaluation Methods Using Equivalent SDOF Model and Elasto-Plastic Response Spectrum

Two simplified methods for evaluating seismic margin due to elasto-plastic response were proposed. Generally, elasto-plastic response is evaluated by nonlinear time-history response analysis using three-dimensional FEM model (3D FEM model). It, however, takes an immense amount of time with commonly used computers. In order to evaluate it in a shorter time, this study developed seismic margin evaluation methods using Equivalent Single Degree Of Freedom (ESDOF) model and elasto-plastic response spectrum. Additionally, the accuracy of the two methods was verified by static loading tests and vibration tests. Simple cantilever test specimens with several natural frequencies were used in the vibration tests, and input waves with several frequency characteristics were applied to each vibration test. Response displacement, response acceleration of the test specimens and input acceleration were measured in each vibration test. Maximum displacement given by ESDOF model of the test specimens was compared with the corresponding measured values of each vibration test in order to verify the accuracy of ESDOF model. Difference between the maximum displacement given by the ESDOF model and the vibration tests was around 5%, and computation time of the ESDOF model was one-tenth of 3D FEM model of the test specimens. In addition, elasto-plastic response spectrum of input waves in the vibration tests were compared with measured yield accelerations of the specimens in order to verify the accuracy of elasto-plastic response spectrum. Difference between the calculated elasto-plastic response spectrum and the measured yield acceleration of the test specimens was around 10%, and computation time of elasto-plastic response spectrum was one-tenth of the 3D FEM model. As a result, it is concluded that ESDOF model and elasto-plastic response spectrum are powerful tool to evaluate seismic margin.Copyright

Detailed and Simplified Approaches for Inelastic Seismic Analysis of Piping Systems

We examined two approaches for predicting the inelastic responses of piping systems. The first enabled us to evaluate accurate responses using nonlinear time-history analysis. We focused on a material model in this approach. Plastic deformation might exhibit a cyclic-hardening effect, which is dependent on the strain ranges, and this hardening affects the dynamic responses of piping systems. Therefore, we improved the hardening model developed by Ohno and Wang (OW model), which is one of the most sophisticated models for producing kinematic hardening behavior. We modified it to produce cyclic-hardening behavior dependent on strain ranges. The second was a simplified approach to evaluate inelastic responses without time-consuming time-history analysis. We developed a tool using equivalent linearization. The tool used analysis techniques including a model of elastic beam elements using flexibility factors for pipe elbows, and modal response spectrum analysis. Equivalent linearization made it possible to apply modal analysis to inelastic analysis. We demonstrated how applicable the approaches were by conducting test simulations.Copyright